{"entries":[{"agent":"Codex","agent_version":"0.164.0","effort":"max","id":"2026-10-31-local-1233340bbd-deepseek-v4.1-flash-1017-max","model":"deepseek/deepseek-v4.1-flash-1017","name":"DeepSeek V4.1 Flash 1017","provider":"OpenRouter","release_date":"2026-10-17","release_date_source":"https://openrouter.ai/deepseek/deepseek-v4.1-flash-1017"},{"agent":"Codex","agent_version":"0.164.0","effort":"max","id":"2026-10-31-local-1233340bbd-gpt-6-luna-max","model":"openai/gpt-6-luna","name":"GPT-6 Luna","provider":"OpenRouter","release_date":"2026-09-22","release_date_source":"https://openrouter.ai/openai/gpt-6-luna"},{"agent":"Codex","agent_version":"0.164.0","effort":"high","id":"2026-10-31-local-1233340bbd-mimo-v2.6-flash-high","model":"xiaomi/mimo-v2.6-flash","name":"MiMo-V2.6-Flash","provider":"OpenRouter","release_date":"2026-09-21","release_date_source":"https://openrouter.ai/xiaomi/mimo-v2.6-flash"}],"id":"2026-11-04-selected-ed4cca047f","observed_at":"2026-10-31T01:44:37.351871+08:00","schema":1,"selection":{"excluded_models":["z-ai/glm-5.3-flash"],"schema":1,"updated_at":"2026-11-04"},"source_hashes":{"index":"d4713073ec3bdc03fb0ae7fe94da09837bde58f461f740b3ebaa250ac522f15d","model_metadata":"1dd1c881a6651c080d617c283d9ea58785eb7a18172f71c0d1924efad56e3dbf","results":"271976ef8c4ea65df18e7dae07ad73aa02b00ee7d26f7d45dbea2a2f707f6ea9"},"tasks":[{"hardware":"cpu","name":"agentic-search","title":"Agentic Search"},{"hardware":"cpu","name":"agentic-search-vectorless","title":"Agentic Search — Vectorless"},{"hardware":"cpu","name":"biomedical-graph-retrieval","title":"Biomedical Graph Retrieval"},{"hardware":"cpu","name":"budgeted-bimetric-retrieval","title":"Budgeted Bimetric Retrieval"},{"hardware":"gpu","name":"budgeted-candidate-reranking","title":"Budgeted Candidate Reranking"},{"hardware":"gpu","name":"chunqiu-temporal-retrieval","title":"ChunQiu Temporal Retrieval"},{"hardware":"cpu","name":"code-context-retrieval","title":"Code Context Retrieval"},{"hardware":"gpu","name":"code-embedding-finetuning","title":"Code Embedding Fine-tuning"},{"hardware":"gpu","name":"code-reranker-finetuning","title":"Code Reranker Fine-tuning"},{"hardware":"gpu","name":"composed-image-search","title":"Composed Image Search"},{"hardware":"gpu","name":"constrained-molecular-search","title":"Constrained Molecular Search"},{"hardware":"gpu","name":"decision-model-training","title":"Decision Model Training"},{"hardware":"gpu","name":"document-page-search","title":"Document Page Search"},{"hardware":"cpu","name":"evolving-personal-memory","title":"Evolving Personal Memory"},{"hardware":"cpu","name":"filtered-ann-retrieval","title":"Filtered ANN Retrieval"},{"hardware":"cpu","name":"financial-report-qa","title":"Financial-Report Evidence Retrieval"},{"hardware":"cpu","name":"icsi-meeting-memory","title":"ICSI Meeting Memory"},{"hardware":"cpu","name":"incremental-vector-search","title":"Incremental Vector Search"},{"hardware":"gpu","name":"instruction-following-embedding","title":"Instruction-following Embedding"},{"hardware":"cpu","name":"join-aware-table-retrieval","title":"Join-aware Table Retrieval"},{"hardware":"gpu","name":"khmer-embedding-training","title":"Khmer-to-English Embedding Training"},{"hardware":"cpu","name":"large-scale-formula-retrieval","title":"Large-scale Formula Retrieval"},{"hardware":"gpu","name":"layout-page-search","title":"Layout Page Search"},{"hardware":"gpu","name":"legal-stance-retrieval","title":"Legal Stance Retrieval"},{"hardware":"gpu","name":"lightweight-query-retrieval","title":"Lightweight Query Retrieval"},{"hardware":"cpu","name":"long-document-reranking","title":"Long-document Reranking"},{"hardware":"cpu","name":"long-pdf-evidence","title":"Long-PDF Sentence Evidence"},{"hardware":"cpu","name":"long-pdf-evidence-vectorless","title":"Long-PDF Sentence Evidence — Vectorless"},{"hardware":"cpu","name":"memory-constrained-dense-retrieval","title":"Memory-constrained Dense Retrieval"},{"hardware":"gpu","name":"methodology-inspiration-retrieval","title":"Methodology Inspiration Retrieval"},{"hardware":"gpu","name":"multilingual-vector-postprocessing","title":"Multilingual Vector Postprocessing"},{"hardware":"cpu","name":"narrative-clue-search","title":"Narrative-Clue Search"},{"hardware":"cpu","name":"negative-constraint-retrieval","title":"Negative-Constraint Retrieval"},{"hardware":"cpu","name":"paper-metadata-retrieval","title":"Paper Metadata Retrieval"},{"hardware":"gpu","name":"protein-text-bidirectional-retrieval","title":"Protein–Text Bidirectional Retrieval"},{"hardware":"gpu","name":"quantity-aware-retrieval","title":"Quantity-aware Retrieval"},{"hardware":"gpu","name":"query-encoder-alignment","title":"Query-side Encoder Alignment"},{"hardware":"cpu","name":"reasoning-query-rewriting","title":"Reasoning-intensive Query Rewriting"},{"hardware":"cpu","name":"rust-bm25-search","title":"Rust BM25 Search"},{"hardware":"gpu","name":"same-place-image-search","title":"Same-Place Image Search"},{"hardware":"cpu","name":"scientific-paper-qa","title":"Scientific-Paper Question Answering"},{"hardware":"gpu","name":"sketch-text-search","title":"Sketch-and-Text Image Search"},{"hardware":"cpu","name":"sparse-retrieval-pruning","title":"Sparse Retrieval Pruning"},{"hardware":"gpu","name":"spoken-query-search","title":"Spoken-Query Document Search"},{"hardware":"gpu","name":"training-error-tracing","title":"Training Error Tracing"},{"hardware":"gpu","name":"video-moment-search","title":"Video-Moment Search"},{"hardware":"cpu","name":"visual-document-qa","title":"Visual-Document Question Answering"}],"trials":[{"agent_seconds":7203.0068,"agent_status":"agent_timeout","configuration":{"agent":"Codex","agent_timeout_sec":7200.0,"coding_commit":"bc2ef55cf5bc4e5133c32816a7874368192a2f1e","effort":"max","model":"deepseek/deepseek-v4.1-flash-1017","provider":"OpenRouter","resources":{"cpus":32,"gpus":0,"memory_mb":16384,"storage_mb":204800},"run_mode":"official_harbor_regrade","task_commit":"ec2e29eddd79d79c00fbe52ea167523911a7514c","verifier_commit":"ec2e29eddd79d79c00fbe52ea167523911a7514c","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-deepseek-v4.1-flash-1017-max","finished_at":"2026-10-25T09:18:02.664741Z","id":"agentic-search-deepseek-v4.1-flash-1017-max-4e4090bad692","redacted_fields":0,"reward":{"gold_recall_at_5":0.0,"jailbreak_judge":1.0,"reward":0.0},"reward_sha256":"7331ad86e2f818fc39e34fce768a0302d8d54170c30536690b04ab02c2af51e2","score":0.0,"started_at":"2026-10-25T07:17:59.657941Z","steps":666,"task":"agentic-search","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":77850028,"input_tokens":78190365,"output_tokens":130916,"reasoning_output_tokens":60050,"response_count":662,"total_tokens":78321281},"trajectory_sha256":"5728123a976dfaacc4e9ea8abb76da763c6cdabdeba039b014df3a0f4ab7809b","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/agentic-search-deepseek-v4.1-flash-1017-max-4e4090bad692/manifest.json","valid":true,"cost_usd":0.363150234},{"agent_seconds":7201.867456,"agent_status":"agent_timeout","configuration":{"agent":"Codex","agent_timeout_sec":7200.0,"coding_commit":"bc2ef55cf5bc4e5133c32816a7874368192a2f1e","effort":"max","model":"openai/gpt-6-luna","provider":"OpenRouter","resources":{"cpus":32,"gpus":0,"memory_mb":16384,"storage_mb":204800},"run_mode":"official_harbor_regrade","task_commit":"ec2e29eddd79d79c00fbe52ea167523911a7514c","verifier_commit":"ec2e29eddd79d79c00fbe52ea167523911a7514c","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-gpt-6-luna-max","finished_at":"2026-10-25T09:18:19.273588Z","id":"agentic-search-gpt-6-luna-max-fd79296ab48e","redacted_fields":0,"reward":{"gold_recall_at_5":0.01,"jailbreak_judge":1.0,"reward":0.01},"reward_sha256":"d5868c0536b74d15a91e3d8e46855065649c9dad381e673d6e86f2e85fd3b058","score":1.0,"started_at":"2026-10-25T07:18:17.406132Z","steps":280,"task":"agentic-search","tokens":{"cache_write_input_tokens":416241,"cached_input_tokens":69124471,"input_tokens":69569763,"output_tokens":295470,"reasoning_output_tokens":209932,"response_count":276,"total_tokens":69865233},"trajectory_sha256":"9d132dba5519633a104e5f1a0c532a527b6073488456f930de6e4dc913fc1a4e","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/agentic-search-gpt-6-luna-max-fd79296ab48e/manifest.json","valid":true,"cost_usd":1.40834168},{"agent_seconds":6859.7007,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":7200.0,"coding_commit":"bc2ef55cf5bc4e5133c32816a7874368192a2f1e","effort":"high","model":"xiaomi/mimo-v2.6-flash","provider":"OpenRouter","resources":{"cpus":32,"gpus":0,"memory_mb":16384,"storage_mb":204800},"run_mode":"official_harbor_regrade","task_commit":"ec2e29eddd79d79c00fbe52ea167523911a7514c","verifier_commit":"ec2e29eddd79d79c00fbe52ea167523911a7514c","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-mimo-v2.6-flash-high","finished_at":"2026-10-25T09:12:19.791978Z","id":"agentic-search-mimo-v2.6-flash-high-343e88dc5f55","redacted_fields":0,"reward":{"gold_recall_at_5":0.1166666667,"jailbreak_judge":1.0,"reward":0.1166666667},"reward_sha256":"a10749d5a7395a006709eca9907ed98ff4270394d0f02a84092e6a6125c47363","score":11.66666667,"started_at":"2026-10-25T07:18:00.091278Z","steps":269,"task":"agentic-search","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":30654688,"input_tokens":31136619,"output_tokens":93416,"reasoning_output_tokens":34706,"response_count":266,"total_tokens":31230035},"trajectory_sha256":"4213e13d397baae4f6f9c806e246412743f5f8131cf1c4d27fbe171849bb9ce0","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/agentic-search-mimo-v2.6-flash-high-343e88dc5f55/manifest.json","valid":true,"cost_usd":0.1794599464},{"agent_seconds":7201.77144,"agent_status":"agent_timeout","configuration":{"agent":"Codex","agent_timeout_sec":7200.0,"coding_commit":"bc2ef55cf5bc4e5133c32816a7874368192a2f1e","effort":"max","model":"deepseek/deepseek-v4.1-flash-1017","provider":"OpenRouter","resources":{"cpus":32,"gpus":0,"memory_mb":16384,"storage_mb":204800},"run_mode":"official_harbor_regrade","task_commit":"0d771a874007e0383a4ba9cc8e8c8d1ca2c63b73","verifier_commit":"82a1fae886305182b8f01c9bbb16e1e654aee726","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-deepseek-v4.1-flash-1017-max","finished_at":"2026-10-24T17:43:16.114224Z","id":"agentic-search-vectorless-deepseek-v4.1-flash-1017-max-158242a28958","redacted_fields":0,"reward":{"gold_recall_at_5":0.0,"jailbreak_judge":1.0,"reward":0.0},"reward_sha256":"5576f7a92564fc035a8371748dfd82518f033bdc6a8e6c3a3e7891aa6677805a","score":0.0,"started_at":"2026-10-24T15:43:14.342784Z","steps":148,"task":"agentic-search-vectorless","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":10971392,"input_tokens":13714128,"output_tokens":97011,"reasoning_output_tokens":31316,"response_count":144,"total_tokens":13811139},"trajectory_sha256":"ab0fb1a27d2b8a5a85e5d5c7eadc776c53421bad47fef325482fab4e359d56b0","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/agentic-search-vectorless-deepseek-v4.1-flash-1017-max-158242a28958/manifest.json","valid":true,"cost_usd":0.502531176},{"agent_seconds":4259.355601,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":7200.0,"coding_commit":"bc2ef55cf5bc4e5133c32816a7874368192a2f1e","effort":"max","model":"openai/gpt-6-luna","provider":"OpenRouter","resources":{"cpus":32,"gpus":0,"memory_mb":16384,"storage_mb":204800},"run_mode":"official_harbor_regrade","task_commit":"82a1fae886305182b8f01c9bbb16e1e654aee726","verifier_commit":"82a1fae886305182b8f01c9bbb16e1e654aee726","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-gpt-6-luna-max","finished_at":"2026-10-24T22:56:23.172130Z","id":"agentic-search-vectorless-gpt-6-luna-max-68163f93f018","redacted_fields":0,"reward":{"gold_recall_at_5":0.0,"jailbreak_judge":1.0,"reward":0.0},"reward_sha256":"5576f7a92564fc035a8371748dfd82518f033bdc6a8e6c3a3e7891aa6677805a","score":0.0,"started_at":"2026-10-24T21:45:23.816529Z","steps":207,"task":"agentic-search-vectorless","tokens":{"cache_write_input_tokens":301780,"cached_input_tokens":36019370,"input_tokens":36321762,"output_tokens":229594,"reasoning_output_tokens":169042,"response_count":204,"total_tokens":36551356},"trajectory_sha256":"afddc830a2d9cc4743781fa9efd52060d5dfb312880d01f6db29afe2d9c168c4","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/agentic-search-vectorless-gpt-6-luna-max-68163f93f018/manifest.json","valid":true,"cost_usd":0.60269973},{"agent_seconds":2731.100365,"agent_status":"output_limit","configuration":{"agent":"Codex","agent_timeout_sec":7200.0,"coding_commit":"bc2ef55cf5bc4e5133c32816a7874368192a2f1e","effort":"high","model":"xiaomi/mimo-v2.6-flash","provider":"OpenRouter","resources":{"cpus":32,"gpus":0,"memory_mb":16384,"storage_mb":204800},"run_mode":"official_harbor_regrade","task_commit":"0d771a874007e0383a4ba9cc8e8c8d1ca2c63b73","verifier_commit":"82a1fae886305182b8f01c9bbb16e1e654aee726","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-mimo-v2.6-flash-high","finished_at":"2026-10-24T16:28:45.280184Z","id":"agentic-search-vectorless-mimo-v2.6-flash-high-14262e34b71b","redacted_fields":0,"reward":{"gold_recall_at_5":0.0,"jailbreak_judge":1.0,"reward":0.0},"reward_sha256":"5576f7a92564fc035a8371748dfd82518f033bdc6a8e6c3a3e7891aa6677805a","score":0.0,"started_at":"2026-10-24T15:43:14.179819Z","steps":19,"task":"agentic-search-vectorless","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":698023,"input_tokens":1396039,"output_tokens":405674,"reasoning_output_tokens":5393,"response_count":16,"total_tokens":1801713},"trajectory_sha256":"0480120ac067eedc5dd423a80c7c951b4c9d0f35bdc759c6789222c7634a2b22","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/agentic-search-vectorless-mimo-v2.6-flash-high-14262e34b71b/manifest.json","valid":true,"cost_usd":0.2132654244},{"agent_seconds":6260.03562,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":7200.0,"coding_commit":null,"effort":"max","model":"deepseek/deepseek-v4.1-flash-1017","provider":"OpenRouter","resources":null,"run_mode":"full_agent_and_verifier","task_commit":null,"verifier_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-deepseek-v4.1-flash-1017-max","finished_at":"2026-10-24T04:56:19.895650Z","id":"biomedical-graph-retrieval-deepseek-v4.1-flash-1017-max-dc1355b97fdc","redacted_fields":0,"reward":{"exact_set_accuracy":0.0,"jailbreak_judge":1.0,"reward":0.0},"reward_sha256":"a82a52709516ffd69da56259559c6fd0a2016b05608572e22e9a36842119d209","score":0.0,"started_at":"2026-10-24T03:11:59.860030Z","steps":405,"task":"biomedical-graph-retrieval","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":82067968,"input_tokens":82338985,"output_tokens":321572,"reasoning_output_tokens":172173,"response_count":402,"total_tokens":82660557},"trajectory_sha256":"75ef9309bf25b1f23a2b0bb17cc853ba58fe5ff1e0ec2842da413e34e54aac91","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/biomedical-graph-retrieval-deepseek-v4.1-flash-1017-max-dc1355b97fdc/manifest.json","valid":true,"cost_usd":0.479799654},{"agent_seconds":2672.684817,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":7200.0,"coding_commit":null,"effort":"max","model":"openai/gpt-6-luna","provider":"OpenRouter","resources":null,"run_mode":"full","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"1537d01c332d4c69160e7106757ac32602ba79b3","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-gpt-6-luna-max","finished_at":"2026-10-24T05:55:07.127116Z","id":"biomedical-graph-retrieval-gpt-6-luna-max-25f203751dbc","redacted_fields":0,"reward":{"exact_set_accuracy":0.1,"jailbreak_judge":1.0,"reward":0.1},"reward_sha256":"8b6ff87d978dcf720bec0ca48e71a473cde610960eae0bc28550689d8149962b","score":10.0,"started_at":"2026-10-24T05:10:34.442299Z","steps":100,"task":"biomedical-graph-retrieval","tokens":{"cache_write_input_tokens":231349,"cached_input_tokens":11146275,"input_tokens":11377906,"output_tokens":177295,"reasoning_output_tokens":80606,"response_count":94,"total_tokens":11555201},"trajectory_sha256":"e401bfb7345760c5569c94134cfb84723879b666f561a5a085093a9a566f2d24","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/biomedical-graph-retrieval-gpt-6-luna-max-25f203751dbc/manifest.json","valid":true,"cost_usd":0.229057075},{"agent_seconds":884.370557,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":7200.0,"coding_commit":null,"effort":"high","model":"xiaomi/mimo-v2.6-flash","provider":"OpenRouter","resources":null,"run_mode":"full","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"1537d01c332d4c69160e7106757ac32602ba79b3","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-mimo-v2.6-flash-high","finished_at":"2026-10-24T04:48:14.960362Z","id":"biomedical-graph-retrieval-mimo-v2.6-flash-high-5b34bffa4066","redacted_fields":0,"reward":{"exact_set_accuracy":0.0,"jailbreak_judge":1.0,"reward":0.0},"reward_sha256":"a82a52709516ffd69da56259559c6fd0a2016b05608572e22e9a36842119d209","score":0.0,"started_at":"2026-10-24T04:33:30.589805Z","steps":47,"task":"biomedical-graph-retrieval","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":2314624,"input_tokens":2723346,"output_tokens":28329,"reasoning_output_tokens":2974,"response_count":44,"total_tokens":2751675},"trajectory_sha256":"841d5fe465aa05f1be025ac743620c85f3044dfe9f8486966e05d5cbf84245f6","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/biomedical-graph-retrieval-mimo-v2.6-flash-high-5b34bffa4066/manifest.json","valid":true,"cost_usd":0.0716341472},{"agent_seconds":4059.628786,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":7200.0,"coding_commit":null,"effort":"max","model":"deepseek/deepseek-v4.1-flash-1017","provider":"OpenRouter","resources":{"cpus":16,"gpus":0,"memory_mb":32768,"storage_mb":65536},"run_mode":"official_harbor_full","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"bc2ef55cf5bc4e5133c32816a7874368192a2f1e","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-deepseek-v4.1-flash-1017-max","finished_at":"2026-10-24T20:33:08.443386Z","id":"budgeted-bimetric-retrieval-deepseek-v4.1-flash-1017-max-0050925e6e09","redacted_fields":0,"reward":{"reward":0.005076703153483426},"reward_sha256":"361ef160db561b74f1734ebb10a53c58b1846ef1f798b8bc48533fe248670903","score":0.5076703153483426,"started_at":"2026-10-24T19:25:28.814600Z","steps":226,"task":"budgeted-bimetric-retrieval","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":21580032,"input_tokens":23005761,"output_tokens":151965,"reasoning_output_tokens":71557,"response_count":223,"total_tokens":23157726},"trajectory_sha256":"7f755d578707e3957f4bc2f2e0e4fde63b6798d3e2ad926b60b9b045b89b3cc1","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/budgeted-bimetric-retrieval-deepseek-v4.1-flash-1017-max-0050925e6e09/manifest.json","valid":true,"cost_usd":0.369778446},{"agent_seconds":2214.668001,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":7200.0,"coding_commit":null,"effort":"max","model":"openai/gpt-6-luna","provider":"OpenRouter","resources":{"cpus":16,"gpus":0,"memory_mb":32768,"storage_mb":65536},"run_mode":"official_harbor_full","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"bc2ef55cf5bc4e5133c32816a7874368192a2f1e","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-gpt-6-luna-max","finished_at":"2026-10-24T21:18:51.920185Z","id":"budgeted-bimetric-retrieval-gpt-6-luna-max-cac411809251","redacted_fields":0,"reward":{"reward":0.18243941745594522},"reward_sha256":"244b95c54a59a60ca0203350f8c957a0ad2bc861c3cb04eb595352ddbbf76508","score":18.243941745594523,"started_at":"2026-10-24T20:41:57.252184Z","steps":99,"task":"budgeted-bimetric-retrieval","tokens":{"cache_write_input_tokens":139462,"cached_input_tokens":7437861,"input_tokens":7577611,"output_tokens":101573,"reasoning_output_tokens":72289,"response_count":96,"total_tokens":7679184},"trajectory_sha256":"3214d3a98b3905932e4cea481b260efcd78e40773b118c0fc128d84a5c7655c6","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/budgeted-bimetric-retrieval-gpt-6-luna-max-cac411809251/manifest.json","valid":true,"cost_usd":0.14262666},{"agent_seconds":1108.530318,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":7200.0,"coding_commit":null,"effort":"high","model":"xiaomi/mimo-v2.6-flash","provider":"OpenRouter","resources":{"cpus":16,"gpus":0,"memory_mb":32768,"storage_mb":65536},"run_mode":"official_harbor_full","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"bc2ef55cf5bc4e5133c32816a7874368192a2f1e","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-mimo-v2.6-flash-high","finished_at":"2026-10-24T21:00:26.267124Z","id":"budgeted-bimetric-retrieval-mimo-v2.6-flash-high-7c348ddd3dc5","redacted_fields":0,"reward":{"reward":0.0},"reward_sha256":"5c73d718100b1b99b632b6c9956dae409fe1fe6b58474aca8d2ea5823bf0c6e4","score":0.0,"started_at":"2026-10-24T20:41:57.736806Z","steps":165,"task":"budgeted-bimetric-retrieval","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":9430336,"input_tokens":9687619,"output_tokens":40195,"reasoning_output_tokens":9435,"response_count":162,"total_tokens":9727814},"trajectory_sha256":"a56fd28ab3f3563145ebfb3ade2ff3ee4606cbf837f2ec3c2812c628685f0a22","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/budgeted-bimetric-retrieval-mimo-v2.6-flash-high-7c348ddd3dc5/manifest.json","valid":true,"cost_usd":0.0736791608},{"agent_seconds":7201.765278,"agent_status":"agent_timeout","configuration":{"agent":"Codex","agent_timeout_sec":7200.0,"coding_commit":null,"effort":"max","model":"deepseek/deepseek-v4.1-flash-1017","provider":"OpenRouter","resources":null,"run_mode":"full","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"1537d01c332d4c69160e7106757ac32602ba79b3","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-deepseek-v4.1-flash-1017-max","finished_at":"2026-10-24T06:21:14.797417Z","id":"budgeted-candidate-reranking-deepseek-v4.1-flash-1017-max-5d8a6d3779e1","redacted_fields":0,"reward":{"reward":0.17134606513982498},"reward_sha256":"91be0bb03cc04dd630763dd48c884d2c25521cd01a4812c548bfc920360e2f4a","score":17.134606513982497,"started_at":"2026-10-24T04:21:13.032139Z","steps":361,"task":"budgeted-candidate-reranking","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":60500864,"input_tokens":60927610,"output_tokens":237776,"reasoning_output_tokens":95907,"response_count":359,"total_tokens":61165386},"trajectory_sha256":"a6c3c65d30d8e6b8fc5ed75816d4a8d98b60f0041d9a7fb5d107634ed12027e5","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/budgeted-candidate-reranking-deepseek-v4.1-flash-1017-max-5d8a6d3779e1/manifest.json","valid":true,"cost_usd":0.388180092},{"agent_seconds":2209.940119,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":7200.0,"coding_commit":null,"effort":"max","model":"openai/gpt-6-luna","provider":"OpenRouter","resources":null,"run_mode":"full","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"1537d01c332d4c69160e7106757ac32602ba79b3","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-gpt-6-luna-max","finished_at":"2026-10-24T05:53:47.318429Z","id":"budgeted-candidate-reranking-gpt-6-luna-max-59c043707509","redacted_fields":0,"reward":{"reward":0.07310787723369022},"reward_sha256":"7ef139b459d50cc1004a524b6f5b21b71c0a469668703a068be6a797534c9129","score":7.310787723369022,"started_at":"2026-10-24T05:16:57.378310Z","steps":104,"task":"budgeted-candidate-reranking","tokens":{"cache_write_input_tokens":165676,"cached_input_tokens":10312377,"input_tokens":10478347,"output_tokens":85487,"reasoning_output_tokens":61430,"response_count":98,"total_tokens":10563834},"trajectory_sha256":"7e138d6a4e24ab6165071b98a61800c71f41e4da210f2814e035a377cadbe06f","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/budgeted-candidate-reranking-gpt-6-luna-max-59c043707509/manifest.json","valid":true,"cost_usd":0.16660617},{"agent_seconds":878.603889,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":7200.0,"coding_commit":null,"effort":"high","model":"xiaomi/mimo-v2.6-flash","provider":"OpenRouter","resources":null,"run_mode":"full","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"1537d01c332d4c69160e7106757ac32602ba79b3","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-mimo-v2.6-flash-high","finished_at":"2026-10-24T05:25:14.096132Z","id":"budgeted-candidate-reranking-mimo-v2.6-flash-high-9a8fd3c0a378","redacted_fields":0,"reward":{"reward":5.464604135575955e-05},"reward_sha256":"edc288308c17a48de5c1981e97767cd940ee875e385c9067e6d7feeee9be1846","score":0.005464604135575955,"started_at":"2026-10-24T05:10:35.492243Z","steps":79,"task":"budgeted-candidate-reranking","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":4449152,"input_tokens":4662967,"output_tokens":25651,"reasoning_output_tokens":3971,"response_count":76,"total_tokens":4688618},"trajectory_sha256":"b7d905ec9631d77ec8365c70be8bdfae109ce79eaf3b946744745f0493b45a9a","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/budgeted-candidate-reranking-mimo-v2.6-flash-high-9a8fd3c0a378/manifest.json","valid":true,"cost_usd":0.0495740056},{"agent_seconds":5588.856035,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":null,"coding_commit":null,"effort":"max","model":"deepseek/deepseek-v4.1-flash-1017","provider":"OpenRouter","resources":null,"run_mode":"saved_outputs_production_rescore","task_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","verifier_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-deepseek-v4.1-flash-1017-max","finished_at":"2026-10-24T08:36:34.762425Z","id":"chunqiu-temporal-retrieval-deepseek-v4.1-flash-1017-max-b1b2c328e060","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"ndcg_at_10":0.004305928768902842,"reward":0.004305928768902842},"reward_sha256":"c69f2bb81a5233230fe7687a00a109246d3e074c1c82631295884cdc8b1ebd76","score":0.4305928768902842,"started_at":"2026-10-24T07:03:25.906390Z","steps":296,"task":"chunqiu-temporal-retrieval","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":65267152,"input_tokens":66484650,"output_tokens":228777,"reasoning_output_tokens":112628,"response_count":293,"total_tokens":66713427},"trajectory_sha256":"9fb0ac9a84fdf83907137f60f6b45252a32493a5bb6a12a3af7d5deb7c39ec0f","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/chunqiu-temporal-retrieval-deepseek-v4.1-flash-1017-max-b1b2c328e060/manifest.json","valid":true,"cost_usd":0.515692356},{"agent_seconds":2958.609505,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":null,"coding_commit":null,"effort":"max","model":"openai/gpt-6-luna","provider":"OpenRouter","resources":null,"run_mode":"saved_outputs_production_rescore","task_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","verifier_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-gpt-6-luna-max","finished_at":"2026-10-24T08:11:07.424679Z","id":"chunqiu-temporal-retrieval-gpt-6-luna-max-d1be8212ebca","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"ndcg_at_10":0.016828333677053258,"reward":0.016828333677053258},"reward_sha256":"178436c258606735f96b0262919e7f0d7898639c9d8feeb980ad3450e3f82e92","score":1.6828333677053258,"started_at":"2026-10-24T07:21:48.815174Z","steps":86,"task":"chunqiu-temporal-retrieval","tokens":{"cache_write_input_tokens":274340,"cached_input_tokens":13278544,"input_tokens":13553651,"output_tokens":184124,"reasoning_output_tokens":135098,"response_count":83,"total_tokens":13737775},"trajectory_sha256":"4c6b521c1a3ef7b1bf294ef7ba69f5a0b061a255a1e04bd3e7ddf2255d82d32d","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/chunqiu-temporal-retrieval-gpt-6-luna-max-d1be8212ebca/manifest.json","valid":true,"cost_usd":0.275167205},{"agent_seconds":1557.514238,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":null,"coding_commit":null,"effort":"high","model":"xiaomi/mimo-v2.6-flash","provider":"OpenRouter","resources":null,"run_mode":"saved_outputs_production_rescore","task_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","verifier_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-mimo-v2.6-flash-high","finished_at":"2026-10-24T07:29:27.334810Z","id":"chunqiu-temporal-retrieval-mimo-v2.6-flash-high-2f65bd67bc82","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"ndcg_at_10":0.0,"reward":0.0},"reward_sha256":"2d7d60fa5b599f75666eacc270a7ec6d84e7e010547b164602ffee3119427b88","score":0.0,"started_at":"2026-10-24T07:03:29.820572Z","steps":68,"task":"chunqiu-temporal-retrieval","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":4202688,"input_tokens":4497978,"output_tokens":42559,"reasoning_output_tokens":3007,"response_count":65,"total_tokens":4540537},"trajectory_sha256":"c5a5aa9fc6bf7a91b201caee88dcad5785ae0bf88ac6f956f5db12c3b5bb984d","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/chunqiu-temporal-retrieval-mimo-v2.6-flash-high-2f65bd67bc82/manifest.json","valid":true,"cost_usd":0.0650246464},{"agent_seconds":7164.204449,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":null,"coding_commit":null,"effort":"max","model":"deepseek/deepseek-v4.1-flash-1017","provider":"OpenRouter","resources":null,"run_mode":"saved_outputs_production_rescore","task_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","verifier_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-deepseek-v4.1-flash-1017-max","finished_at":"2026-10-24T11:32:06.891463Z","id":"code-context-retrieval-deepseek-v4.1-flash-1017-max-8429aebfcf01","redacted_fields":0,"reward":{"line_f1":0.16741224334783394,"reward":0.16741224334783394},"reward_sha256":"a31802f2e06131a967599408478d88d4b890cf506c107cfa19480e352e87b595","score":16.741224334783393,"started_at":"2026-10-24T09:32:42.687014Z","steps":242,"task":"code-context-retrieval","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":53845504,"input_tokens":54729695,"output_tokens":211435,"reasoning_output_tokens":87060,"response_count":239,"total_tokens":54941130},"trajectory_sha256":"0fa7b8d44614dff8ea17114604e2fc2c86de1ba03433d8cab936f7d15c89c59f","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/code-context-retrieval-deepseek-v4.1-flash-1017-max-8429aebfcf01/manifest.json","valid":true,"cost_usd":0.421026162},{"agent_seconds":5718.594794,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":null,"coding_commit":null,"effort":"max","model":"openai/gpt-6-luna","provider":"OpenRouter","resources":null,"run_mode":"saved_outputs_production_rescore","task_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","verifier_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-gpt-6-luna-max","finished_at":"2026-10-24T11:11:48.067305Z","id":"code-context-retrieval-gpt-6-luna-max-11c1e408bc2a","redacted_fields":0,"reward":{"line_f1":0.16710384666780304,"reward":0.16710384666780304},"reward_sha256":"c79b47cb3fca53e7c97d8a827e4f36a7cd7af58fadd9868b8523a4c31fe840a9","score":16.710384666780303,"started_at":"2026-10-24T09:36:29.472511Z","steps":175,"task":"code-context-retrieval","tokens":{"cache_write_input_tokens":463014,"cached_input_tokens":51348852,"input_tokens":51839738,"output_tokens":353680,"reasoning_output_tokens":268145,"response_count":172,"total_tokens":52193418},"trajectory_sha256":"e84d4249dd729e00b880b781196368ca7619d219995c52551e739681f78245b5","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/code-context-retrieval-gpt-6-luna-max-11c1e408bc2a/manifest.json","valid":true,"cost_usd":1.22600387},{"agent_seconds":3250.080505,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":null,"coding_commit":null,"effort":"high","model":"xiaomi/mimo-v2.6-flash","provider":"OpenRouter","resources":null,"run_mode":"saved_outputs_production_rescore","task_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","verifier_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-mimo-v2.6-flash-high","finished_at":"2026-10-24T10:26:52.345612Z","id":"code-context-retrieval-mimo-v2.6-flash-high-e2e50bf629d2","redacted_fields":0,"reward":{"line_f1":0.01575064049029029,"reward":0.01575064049029029},"reward_sha256":"9cca51dca5d4cedfe7b448498f751e7e610513822840f055613e39e3e395c7df","score":1.575064049029029,"started_at":"2026-10-24T09:32:42.265107Z","steps":193,"task":"code-context-retrieval","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":14164672,"input_tokens":15242970,"output_tokens":55220,"reasoning_output_tokens":11116,"response_count":190,"total_tokens":15298190},"trajectory_sha256":"d3272200ef07be1891c7bb4e184771eee8af35cf3465951f84447f061bad9205","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/code-context-retrieval-mimo-v2.6-flash-high-e2e50bf629d2/manifest.json","valid":true,"cost_usd":0.2060844016},{"agent_seconds":17902.595126,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":null,"coding_commit":null,"effort":"max","model":"deepseek/deepseek-v4.1-flash-1017","provider":"OpenRouter","resources":null,"run_mode":"official_harbor_full","task_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","verifier_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-deepseek-v4.1-flash-1017-max","finished_at":"2026-10-28T16:30:43.189254Z","id":"code-embedding-finetuning-deepseek-v4.1-flash-1017-max-695008502793","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"retrieval_improvement":0.17,"reward":0.17},"reward_sha256":"658d3509fb7713086320abc78a74d0ddefdf0e4e1cfd28a72070e900b5242c19","score":17.0,"started_at":"2026-10-28T11:32:20.594128Z","steps":467,"task":"code-embedding-finetuning","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":72477442,"input_tokens":74271355,"output_tokens":235389,"reasoning_output_tokens":136528,"response_count":464,"total_tokens":74506744},"trajectory_sha256":"7871e515c4f603a907b8665e8ebc1faebd3ceae59ed9f00574a4319f9ae661e5","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/code-embedding-finetuning-deepseek-v4.1-flash-1017-max-695008502793/manifest.json","valid":true,"cost_usd":0.627752676},{"agent_seconds":10049.518373,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":null,"coding_commit":null,"effort":"max","model":"openai/gpt-6-luna","provider":"OpenRouter","resources":null,"run_mode":"official_harbor_regrade","task_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","verifier_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-gpt-6-luna-max","finished_at":"2026-10-28T14:19:51.060625Z","id":"code-embedding-finetuning-gpt-6-luna-max-8bccb3833ae3","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"retrieval_improvement":0.05,"reward":0.05},"reward_sha256":"099f733559cf07e2bc13e2b2afea72b41a9941343793ac6c9f53cead7a7dc0cf","score":5.0,"started_at":"2026-10-28T11:32:21.542252Z","steps":164,"task":"code-embedding-finetuning","tokens":{"cache_write_input_tokens":200655,"cached_input_tokens":18996060,"input_tokens":19197198,"output_tokens":103161,"reasoning_output_tokens":72439,"response_count":161,"total_tokens":19300359},"trajectory_sha256":"ca79f4e20ec97178c31966cbf94d63959ccf91279ca3fa83b99b62347bc386ba","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/code-embedding-finetuning-gpt-6-luna-max-8bccb3833ae3/manifest.json","valid":true,"cost_usd":0.266671275},{"agent_seconds":5408.670106,"agent_status":"output_limit","configuration":{"agent":"Codex","agent_timeout_sec":null,"coding_commit":null,"effort":"high","model":"xiaomi/mimo-v2.6-flash","provider":"OpenRouter","resources":null,"run_mode":"official_harbor_full","task_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","verifier_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-mimo-v2.6-flash-high","finished_at":"2026-10-28T13:02:30.574972Z","id":"code-embedding-finetuning-mimo-v2.6-flash-high-c295ca35f645","redacted_fields":0,"reward":{"jailbreak_judge":0,"retrieval_improvement":0,"reward":0},"reward_sha256":"f94a543a75d5fab2062620f09adbb54df48eeec1b285a76d614312182ca5d444","score":0.0,"started_at":"2026-10-28T11:32:21.904866Z","steps":31,"task":"code-embedding-finetuning","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":696721,"input_tokens":2968279,"output_tokens":548063,"reasoning_output_tokens":1423,"response_count":28,"total_tokens":3516342},"trajectory_sha256":"a1a0ba4e2886b6eb14e110d023309e607882d2c7f5cb74eba13dfcd8ae64e313","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/code-embedding-finetuning-mimo-v2.6-flash-high-c295ca35f645/manifest.json","valid":true,"cost_usd":0.4734265788},{"agent_seconds":2334.547588,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":21600.0,"coding_commit":null,"effort":"max","model":"deepseek/deepseek-v4.1-flash-1017","provider":"OpenRouter","resources":{"cpus":8,"gpus":1,"memory_mb":32768,"storage_mb":102400},"run_mode":"official_harbor_full","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"a1ede723abe0add54045301f49a187e11339c134","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-deepseek-v4.1-flash-1017-max","finished_at":"2026-10-24T11:11:00.177852Z","id":"code-reranker-finetuning-deepseek-v4.1-flash-1017-max-6f67dc848c7a","redacted_fields":0,"reward":{"reward":0.0},"reward_sha256":"1d8511dcb533d7d0ea095ab1fed80023c526956199da238a47b3bcde6e6df655","score":0.0,"started_at":"2026-10-24T10:32:05.630264Z","steps":79,"task":"code-reranker-finetuning","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":3200384,"input_tokens":3470080,"output_tokens":64228,"reasoning_output_tokens":48911,"response_count":77,"total_tokens":3534308},"trajectory_sha256":"c0a3b9fe6a5489b03bab5b2b5b9e108709c0c1e4a4c73cb5873ba5aa2b42249a","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/code-reranker-finetuning-deepseek-v4.1-flash-1017-max-6f67dc848c7a/manifest.json","valid":true,"cost_usd":0.088592352},{"agent_seconds":11220.392465,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":21600.0,"coding_commit":null,"effort":"max","model":"openai/gpt-6-luna","provider":"OpenRouter","resources":{"cpus":8,"gpus":1,"memory_mb":32768,"storage_mb":102400},"run_mode":"official_harbor_full","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"bc2ef55cf5bc4e5133c32816a7874368192a2f1e","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-gpt-6-luna-max","finished_at":"2026-10-24T21:11:06.512627Z","id":"code-reranker-finetuning-gpt-6-luna-max-ef96da1ac661","redacted_fields":0,"reward":{"reward":0.4090909090909091},"reward_sha256":"283b698697cc9dcf4258cb788edf753fd18cb3a3c98ef2d4559c94ace2906a6d","score":40.909090909090914,"started_at":"2026-10-24T18:04:06.120162Z","steps":143,"task":"code-reranker-finetuning","tokens":{"cache_write_input_tokens":159055,"cached_input_tokens":13725215,"input_tokens":13884690,"output_tokens":91089,"reasoning_output_tokens":66067,"response_count":140,"total_tokens":13975779},"trajectory_sha256":"0ca97bd191fe6317faf5cd57e414d23806f639f7037512cfb097259685edccf6","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/code-reranker-finetuning-gpt-6-luna-max-ef96da1ac661/manifest.json","valid":true,"cost_usd":0.202720525},{"agent_seconds":4656.882207,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":21600.0,"coding_commit":null,"effort":"high","model":"xiaomi/mimo-v2.6-flash","provider":"OpenRouter","resources":{"cpus":8,"gpus":1,"memory_mb":32768,"storage_mb":102400},"run_mode":"official_harbor_full","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"a1ede723abe0add54045301f49a187e11339c134","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-mimo-v2.6-flash-high","finished_at":"2026-10-24T11:49:43.386389Z","id":"code-reranker-finetuning-mimo-v2.6-flash-high-4f6a3a5c7b0f","redacted_fields":0,"reward":{"reward":0.25000000000000006},"reward_sha256":"2822aa72d6209d96bd8b18c7e0b1787adb843a4bbed2fad82be32c474130586d","score":25.000000000000007,"started_at":"2026-10-24T10:32:06.504182Z","steps":173,"task":"code-reranker-finetuning","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":12112320,"input_tokens":15602013,"output_tokens":126135,"reasoning_output_tokens":16466,"response_count":170,"total_tokens":15728148},"trajectory_sha256":"34d213e1320ed645108b28784e2786d18a3f4ce4b8b12ca267c2baaea26c41c5","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/code-reranker-finetuning-mimo-v2.6-flash-high-4f6a3a5c7b0f/manifest.json","valid":true,"cost_usd":0.557789316},{"agent_seconds":7206.712192,"agent_status":"agent_timeout","configuration":{"agent":"Codex","agent_timeout_sec":7200.0,"coding_commit":null,"effort":"max","model":"deepseek/deepseek-v4.1-flash-1017","provider":"OpenRouter","resources":{"cpus":8,"gpus":1,"memory_mb":32768,"storage_mb":102400},"run_mode":"official_harbor_regrade","task_commit":"052045836520335c1caa5e4a06ed887faacef55f","verifier_commit":"052045836520335c1caa5e4a06ed887faacef55f","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-deepseek-v4.1-flash-1017-max","finished_at":"2026-10-26T06:28:57.794172Z","id":"composed-image-search-deepseek-v4.1-flash-1017-max-3aa34ef4cdd7","redacted_fields":0,"reward":{"composed_source_quality":0.355877064,"jailbreak_judge":1.0,"reward":0.2959244583},"reward_sha256":"9ba3c16e341cf262e1df6506fe81be152e1f3c285feba57b4a76d213997e5c06","score":29.592445830000003,"started_at":"2026-10-26T04:28:51.081980Z","steps":176,"task":"composed-image-search","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":13275468,"input_tokens":15141433,"output_tokens":202640,"reasoning_output_tokens":122794,"response_count":172,"total_tokens":15344073},"trajectory_sha256":"b488fa763dd7613c6fdcad79c0383a9a8fcff3601767cdbff6be440979358fc0","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/composed-image-search-deepseek-v4.1-flash-1017-max-3aa34ef4cdd7/manifest.json","valid":true,"cost_usd":0.441305154},{"agent_seconds":2976.757596,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":null,"coding_commit":null,"effort":"max","model":"openai/gpt-6-luna","provider":"OpenRouter","resources":null,"run_mode":"official_harbor_full","task_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","verifier_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-gpt-6-luna-max","finished_at":"2026-10-28T17:49:44.854648Z","id":"composed-image-search-gpt-6-luna-max-4800b458acbd","redacted_fields":0,"reward":{"composed_source_quality":0.4176574462,"jailbreak_judge":1.0,"reward":0.3634551324},"reward_sha256":"e3f13f68b70919454f2866072de8ead8884a2cbd950f2f7f6abe12a73f4996fc","score":36.34551324,"started_at":"2026-10-28T17:00:08.097052Z","steps":64,"task":"composed-image-search","tokens":{"cache_write_input_tokens":182693,"cached_input_tokens":6469145,"input_tokens":6697157,"output_tokens":142581,"reasoning_output_tokens":106553,"response_count":61,"total_tokens":6839738},"trajectory_sha256":"64eee443332d789f200ff35cbec859332c99bed5348ac40c8fe7c8985d215221","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/composed-image-search-gpt-6-luna-max-4800b458acbd/manifest.json","valid":true,"cost_usd":0.163350475},{"agent_seconds":732.900334,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":7200.0,"coding_commit":null,"effort":"high","model":"xiaomi/mimo-v2.6-flash","provider":"OpenRouter","resources":{"cpus":8,"gpus":1,"memory_mb":32768,"storage_mb":102400},"run_mode":"official_harbor_regrade","task_commit":"052045836520335c1caa5e4a06ed887faacef55f","verifier_commit":"052045836520335c1caa5e4a06ed887faacef55f","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-mimo-v2.6-flash-high","finished_at":"2026-10-25T11:54:52.141792Z","id":"composed-image-search-mimo-v2.6-flash-high-66b3343ac644","redacted_fields":0,"reward":{"composed_source_quality":0.0,"jailbreak_judge":1.0,"reward":0.0},"reward_sha256":"6dc98cd31e25a375d284d86bc8bc19e8b5478afb97e1aa3cae12c8a7b4ebc9b0","score":0.0,"started_at":"2026-10-25T11:42:39.241458Z","steps":31,"task":"composed-image-search","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":1252928,"input_tokens":1540080,"output_tokens":47093,"reasoning_output_tokens":1019,"response_count":28,"total_tokens":1587173},"trajectory_sha256":"3916b650394c5f43e5dfb7587bba5f35d87deab512cb809f7c43086b4a0c45b9","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/composed-image-search-mimo-v2.6-flash-high-66b3343ac644/manifest.json","valid":true,"cost_usd":0.0568955184},{"agent_seconds":15436.551561,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":21600.0,"coding_commit":"bc2ef55cf5bc4e5133c32816a7874368192a2f1e","effort":"max","model":"deepseek/deepseek-v4.1-flash-1017","provider":"OpenRouter","resources":{"cpus":16,"gpus":1,"memory_mb":32768,"storage_mb":65536},"run_mode":"official_harbor_regrade","task_commit":"82a1fae886305182b8f01c9bbb16e1e654aee726","verifier_commit":"82a1fae886305182b8f01c9bbb16e1e654aee726","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-deepseek-v4.1-flash-1017-max","finished_at":"2026-10-25T03:44:57.625378Z","id":"constrained-molecular-search-deepseek-v4.1-flash-1017-max-21295b9a727d","redacted_fields":0,"reward":{"reward":0.13625199999999998},"reward_sha256":"090b172f9343e26133cf689def0fb5f1688eb4386993c918614a4fb94d3a0432","score":13.625199999999998,"started_at":"2026-10-24T23:27:41.073817Z","steps":710,"task":"constrained-molecular-search","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":210097152,"input_tokens":214341252,"output_tokens":306148,"reasoning_output_tokens":146076,"response_count":706,"total_tokens":214647400},"trajectory_sha256":"199a1b41303f6eac3fcb0fa1681962167012b111d6a26ac53e4d8e0b925dde41","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/constrained-molecular-search-deepseek-v4.1-flash-1017-max-21295b9a727d/manifest.json","valid":true,"cost_usd":1.450595256},{"agent_seconds":9621.832471,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":21600.0,"coding_commit":"bc2ef55cf5bc4e5133c32816a7874368192a2f1e","effort":"max","model":"openai/gpt-6-luna","provider":"OpenRouter","resources":{"cpus":16,"gpus":1,"memory_mb":32768,"storage_mb":65536},"run_mode":"official_harbor_regrade","task_commit":"82a1fae886305182b8f01c9bbb16e1e654aee726","verifier_commit":"82a1fae886305182b8f01c9bbb16e1e654aee726","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-gpt-6-luna-max","finished_at":"2026-10-24T23:43:29.265554Z","id":"constrained-molecular-search-gpt-6-luna-max-faf64ca3dc77","redacted_fields":0,"reward":{"reward":0.155102},"reward_sha256":"6e05d773a589d232185879d680b69d7a963c08800b08cf63ef5c7c7354c23a6e","score":15.5102,"started_at":"2026-10-24T21:03:07.433083Z","steps":255,"task":"constrained-molecular-search","tokens":{"cache_write_input_tokens":307856,"cached_input_tokens":55410774,"input_tokens":55719386,"output_tokens":200567,"reasoning_output_tokens":156618,"response_count":252,"total_tokens":55919953},"trajectory_sha256":"af87fe7e3de479d6c1eec0ef59de59ba38ae6f2ef10fc8126ed84e8badb413fe","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/constrained-molecular-search-gpt-6-luna-max-faf64ca3dc77/manifest.json","valid":true,"cost_usd":0.975984745},{"agent_seconds":677.82186,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":21600.0,"coding_commit":"bc2ef55cf5bc4e5133c32816a7874368192a2f1e","effort":"high","model":"xiaomi/mimo-v2.6-flash","provider":"OpenRouter","resources":{"cpus":16,"gpus":1,"memory_mb":32768,"storage_mb":65536},"run_mode":"official_harbor_regrade","task_commit":"82a1fae886305182b8f01c9bbb16e1e654aee726","verifier_commit":"82a1fae886305182b8f01c9bbb16e1e654aee726","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-mimo-v2.6-flash-high","finished_at":"2026-10-24T21:37:46.242609Z","id":"constrained-molecular-search-mimo-v2.6-flash-high-886f74114e58","redacted_fields":0,"reward":{"reward":0.08147599999999999},"reward_sha256":"15257021915bcdd1ed76e3692055a8770ec04c13bfa11fe035d6b81e4ff95c48","score":8.147599999999999,"started_at":"2026-10-24T21:26:28.420749Z","steps":38,"task":"constrained-molecular-search","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":2084288,"input_tokens":2470192,"output_tokens":50146,"reasoning_output_tokens":3119,"response_count":35,"total_tokens":2520338},"trajectory_sha256":"bd598437d91fca61acfdd92087a790cf1f4b18058c59dd4fa27c16f7b6d65e02","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/constrained-molecular-search-mimo-v2.6-flash-high-886f74114e58/manifest.json","valid":true,"cost_usd":0.0739034464},{"agent_seconds":1002.626275,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":21600.0,"coding_commit":null,"effort":"max","model":"deepseek/deepseek-v4.1-flash-1017","provider":"OpenRouter","resources":{"cpus":16,"gpus":1,"memory_mb":32768,"storage_mb":102400},"run_mode":"official_harbor_full","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"a1ede723abe0add54045301f49a187e11339c134","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-deepseek-v4.1-flash-1017-max","finished_at":"2026-10-24T09:52:54.004256Z","id":"decision-model-training-deepseek-v4.1-flash-1017-max-a3f9fd70a3ac","redacted_fields":0,"reward":{"accuracy":0,"reward":0},"reward_sha256":"1c7310a716acb805ab97e738fe250259125ad6a731856a63fb85d8a96b575a62","score":0.0,"started_at":"2026-10-24T09:36:11.377981Z","steps":16,"task":"decision-model-training","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":315776,"input_tokens":409275,"output_tokens":147003,"reasoning_output_tokens":47575,"response_count":13,"total_tokens":556278},"trajectory_sha256":"3fd1fec266a2f39301c08aa9accda62110118cd8713b2f431c438fb5656c6aa6","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/decision-model-training-deepseek-v4.1-flash-1017-max-a3f9fd70a3ac/manifest.json","valid":true,"cost_usd":0.103173978},{"agent_seconds":12706.505329,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":21600.0,"coding_commit":null,"effort":"max","model":"openai/gpt-6-luna","provider":"OpenRouter","resources":{"cpus":16,"gpus":1,"memory_mb":32768,"storage_mb":102400},"run_mode":"official_harbor_regrade","task_commit":"bc2ef55cf5bc4e5133c32816a7874368192a2f1e","verifier_commit":"bc2ef55cf5bc4e5133c32816a7874368192a2f1e","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-gpt-6-luna-max","finished_at":"2026-10-25T19:52:16.418735Z","id":"decision-model-training-gpt-6-luna-max-f9719534b2c0","redacted_fields":0,"reward":{"accuracy":0.3826190476190476,"reward":0.3826190476190476},"reward_sha256":"2c1f22d12684ade480dfa2d39e8fb05c5ad2b94d77fbcafcc37faa25b99ae3fa","score":38.26190476190476,"started_at":"2026-10-25T16:20:29.913406Z","steps":150,"task":"decision-model-training","tokens":{"cache_write_input_tokens":218045,"cached_input_tokens":21531216,"input_tokens":21749702,"output_tokens":165556,"reasoning_output_tokens":123555,"response_count":147,"total_tokens":21915258},"trajectory_sha256":"0deef235128635b096b32f9740c075c7ad27bc7946324a17710bc2e1620f90a9","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/decision-model-training-gpt-6-luna-max-f9719534b2c0/manifest.json","valid":true,"cost_usd":0.325389885},{"agent_seconds":2391.4379,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":21600.0,"coding_commit":null,"effort":"high","model":"xiaomi/mimo-v2.6-flash","provider":"OpenRouter","resources":{"cpus":16,"gpus":1,"memory_mb":32768,"storage_mb":102400},"run_mode":"official_harbor_full","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"a1ede723abe0add54045301f49a187e11339c134","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-mimo-v2.6-flash-high","finished_at":"2026-10-24T07:43:20.341935Z","id":"decision-model-training-mimo-v2.6-flash-high-8116b63eae76","redacted_fields":0,"reward":{"accuracy":0,"reward":0},"reward_sha256":"1c7310a716acb805ab97e738fe250259125ad6a731856a63fb85d8a96b575a62","score":0.0,"started_at":"2026-10-24T07:03:28.904035Z","steps":89,"task":"decision-model-training","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":4997440,"input_tokens":5307875,"output_tokens":51768,"reasoning_output_tokens":5948,"response_count":87,"total_tokens":5359643},"trajectory_sha256":"5eacb7c028b848bccee882062d67db41541b70e2a2fcef293a4f45dda20b087c","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/decision-model-training-mimo-v2.6-flash-high-8116b63eae76/manifest.json","valid":true,"cost_usd":0.071948772},{"agent_seconds":7202.340865,"agent_status":"agent_timeout","configuration":{"agent":"Codex","agent_timeout_sec":7200.0,"coding_commit":null,"effort":"max","model":"deepseek/deepseek-v4.1-flash-1017","provider":"OpenRouter","resources":{"cpus":8,"gpus":1,"memory_mb":32768,"storage_mb":102400},"run_mode":"official_harbor_regrade","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"bc2ef55cf5bc4e5133c32816a7874368192a2f1e","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-deepseek-v4.1-flash-1017-max","finished_at":"2026-10-24T16:38:52.893583Z","id":"document-page-search-deepseek-v4.1-flash-1017-max-d9fccc6b44e5","redacted_fields":0,"reward":{"domain_macro_ndcg":0.0,"jailbreak_judge":1.0,"reward":0.0},"reward_sha256":"0f70b15dfbd3def3f25f3a801b3608c5b8fab8fc5f930f81db6eeed2f16fdec7","score":0.0,"started_at":"2026-10-24T14:38:50.552718Z","steps":126,"task":"document-page-search","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":9393920,"input_tokens":10434763,"output_tokens":133430,"reasoning_output_tokens":84452,"response_count":122,"total_tokens":10568193},"trajectory_sha256":"6c82c8e205b19de02cf24df6216781f8e2f0763b5c4b65b4a03df6afbf051dd9","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/document-page-search-deepseek-v4.1-flash-1017-max-d9fccc6b44e5/manifest.json","valid":true,"cost_usd":0.26436621},{"agent_seconds":3791.994709,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":7200.0,"coding_commit":null,"effort":"max","model":"openai/gpt-6-luna","provider":"OpenRouter","resources":{"cpus":8,"gpus":1,"memory_mb":32768,"storage_mb":102400},"run_mode":"official_harbor_full","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"a1ede723abe0add54045301f49a187e11339c134","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-gpt-6-luna-max","finished_at":"2026-10-24T14:44:22.122213Z","id":"document-page-search-gpt-6-luna-max-ba366eb02dc4","redacted_fields":0,"reward":{"domain_macro_ndcg":0.188168564,"jailbreak_judge":1.0,"reward":0.0701776558},"reward_sha256":"b7f523c89fdc44e5377f1469e4ba22cc91a1a5344e356b8b1f52e6a93962df80","score":7.01776558,"started_at":"2026-10-24T13:41:10.127504Z","steps":126,"task":"document-page-search","tokens":{"cache_write_input_tokens":210293,"cached_input_tokens":16319010,"input_tokens":16534852,"output_tokens":134622,"reasoning_output_tokens":97304,"response_count":123,"total_tokens":16669474},"trajectory_sha256":"204decb330e87b35bf543e02eb374c04bc3b11379f11ed7ec2b86ac48defbc58","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/document-page-search-gpt-6-luna-max-ba366eb02dc4/manifest.json","valid":true,"cost_usd":0.257342625},{"agent_seconds":3641.012251,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":7200.0,"coding_commit":null,"effort":"high","model":"xiaomi/mimo-v2.6-flash","provider":"OpenRouter","resources":{"cpus":8,"gpus":1,"memory_mb":32768,"storage_mb":102400},"run_mode":"official_harbor_full","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"a1ede723abe0add54045301f49a187e11339c134","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-mimo-v2.6-flash-high","finished_at":"2026-10-24T13:19:39.200999Z","id":"document-page-search-mimo-v2.6-flash-high-7f4786e13a05","redacted_fields":0,"reward":{"domain_macro_ndcg":0.104584033,"jailbreak_judge":1.0,"reward":0.0390047206},"reward_sha256":"29828851624699d9aeb0da35a83c381403fac5e28601527d85aa9ec5fb2b6a3e","score":3.9004720600000002,"started_at":"2026-10-24T12:18:58.188748Z","steps":197,"task":"document-page-search","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":14976581,"input_tokens":15378494,"output_tokens":68930,"reasoning_output_tokens":7316,"response_count":194,"total_tokens":15447424},"trajectory_sha256":"c3289b9c3b2df79fc4f2d01e34d64e825cc258b749647ad68f59525a524cd3d2","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/document-page-search-mimo-v2.6-flash-high-7f4786e13a05/manifest.json","valid":true,"cost_usd":0.1175026468},{"agent_seconds":2500.570909,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":7200,"coding_commit":null,"effort":"max","model":"deepseek/deepseek-v4.1-flash-1017","provider":"OpenRouter","resources":{"cpus":8,"gpus":0,"memory_mb":32768,"storage_mb":102400},"run_mode":"official_harbor_full","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"bc2ef55cf5bc4e5133c32816a7874368192a2f1e","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-deepseek-v4.1-flash-1017-max","finished_at":"2026-10-24T22:55:02.334795Z","id":"evolving-personal-memory-deepseek-v4.1-flash-1017-max-4e00798ac5fb","redacted_fields":0,"reward":{"fama":0.3627455136,"jailbreak_judge":1.0,"reward":0.195433456},"reward_sha256":"5f381661bf2ddac402690cb68673bceae513e349b5aadd3b20e2f705f54154cf","score":19.543345600000002,"started_at":"2026-10-24T22:13:21.763886Z","steps":182,"task":"evolving-personal-memory","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":54305408,"input_tokens":56260315,"output_tokens":305119,"reasoning_output_tokens":192431,"response_count":179,"total_tokens":56565434},"trajectory_sha256":"addacb2521f42013d635e4e956d6274637168a65b109d1fbd3101da5eb416a6c","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/evolving-personal-memory-deepseek-v4.1-flash-1017-max-4e00798ac5fb/manifest.json","valid":true,"cost_usd":0.639223674},{"agent_seconds":2711.010582,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":7200,"coding_commit":null,"effort":"max","model":"openai/gpt-6-luna","provider":"OpenRouter","resources":{"cpus":8,"gpus":0,"memory_mb":32768,"storage_mb":102400},"run_mode":"official_harbor_full","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"bc2ef55cf5bc4e5133c32816a7874368192a2f1e","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-gpt-6-luna-max","finished_at":"2026-10-24T23:52:43.811008Z","id":"evolving-personal-memory-gpt-6-luna-max-5f053fbb3ab8","redacted_fields":0,"reward":{"fama":0.4716443025,"jailbreak_judge":1.0,"reward":0.3277687373},"reward_sha256":"75467739e5bc69011854cea4b6c2277dbff203993f74e868de869b05cbd26976","score":32.77687373,"started_at":"2026-10-24T23:07:32.800426Z","steps":103,"task":"evolving-personal-memory","tokens":{"cache_write_input_tokens":246986,"cached_input_tokens":14836036,"input_tokens":15087765,"output_tokens":191631,"reasoning_output_tokens":136197,"response_count":100,"total_tokens":15279396},"trajectory_sha256":"a0c2a842f0a274d07b85e98fa49e43f32c44a969b66461d5421bbdc2ba4f6378","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/evolving-personal-memory-gpt-6-luna-max-5f053fbb3ab8/manifest.json","valid":true,"cost_usd":0.27552341},{"agent_seconds":1914.328935,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":7200,"coding_commit":null,"effort":"high","model":"xiaomi/mimo-v2.6-flash","provider":"OpenRouter","resources":{"cpus":8,"gpus":0,"memory_mb":32768,"storage_mb":102400},"run_mode":"official_harbor_full","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"bc2ef55cf5bc4e5133c32816a7874368192a2f1e","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-mimo-v2.6-flash-high","finished_at":"2026-10-24T23:32:35.706578Z","id":"evolving-personal-memory-mimo-v2.6-flash-high-844c5f6956a6","redacted_fields":0,"reward":{"fama":0.151539963,"jailbreak_judge":1.0,"reward":0.0816439558},"reward_sha256":"2a3a108f3768011363c1a4e2c1ea2381d37d1f22f77b1a6abcf3c38109003b3a","score":8.16439558,"started_at":"2026-10-24T23:00:41.377643Z","steps":61,"task":"evolving-personal-memory","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":3666760,"input_tokens":4616822,"output_tokens":311520,"reasoning_output_tokens":14287,"response_count":58,"total_tokens":4928342},"trajectory_sha256":"322bfff7ac4545269bdf7a362a19b4636a30d9b0ab824b406b48804d65e27054","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/evolving-personal-memory-mimo-v2.6-flash-high-844c5f6956a6/manifest.json","valid":true,"cost_usd":0.230501208},{"agent_seconds":7201.857348,"agent_status":"agent_timeout","configuration":{"agent":"Codex","agent_timeout_sec":7200.0,"coding_commit":null,"effort":"max","model":"deepseek/deepseek-v4.1-flash-1017","provider":"OpenRouter","resources":null,"run_mode":"official_harbor_regrade","task_commit":null,"verifier_commit":"a9fa0cc755dbd0929005e0fb30ff454c19d2798f","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-deepseek-v4.1-flash-1017-max","finished_at":"2026-10-24T05:12:02.004023Z","id":"filtered-ann-retrieval-deepseek-v4.1-flash-1017-max-34f3c9665313","redacted_fields":0,"reward":{"filtered_top10":0.0,"jailbreak_judge":1.0,"reward":0.0},"reward_sha256":"29bb22e2f2f6ee057f1338c43daa1102560d1aca0ebc580acd183d66dc3ce772","score":0.0,"started_at":"2026-10-24T03:12:00.146675Z","steps":253,"task":"filtered-ann-retrieval","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":30673024,"input_tokens":31691383,"output_tokens":282144,"reasoning_output_tokens":208295,"response_count":250,"total_tokens":31973527},"trajectory_sha256":"81bcf040a8cec80af978f1ea569fd0a1c8a35b3554d8cbfce6ea1a4eaf067279","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/filtered-ann-retrieval-deepseek-v4.1-flash-1017-max-34f3c9665313/manifest.json","valid":true,"cost_usd":0.414059322},{"agent_seconds":3325.34798,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":7200.0,"coding_commit":null,"effort":"max","model":"openai/gpt-6-luna","provider":"OpenRouter","resources":null,"run_mode":"official_harbor_regrade","task_commit":null,"verifier_commit":"a9fa0cc755dbd0929005e0fb30ff454c19d2798f","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-gpt-6-luna-max","finished_at":"2026-10-24T05:10:40.627135Z","id":"filtered-ann-retrieval-gpt-6-luna-max-224e6cb63522","redacted_fields":0,"reward":{"filtered_top10":0.0,"jailbreak_judge":1.0,"reward":0.0},"reward_sha256":"29bb22e2f2f6ee057f1338c43daa1102560d1aca0ebc580acd183d66dc3ce772","score":0.0,"started_at":"2026-10-24T04:15:15.279155Z","steps":79,"task":"filtered-ann-retrieval","tokens":{"cache_write_input_tokens":196948,"cached_input_tokens":9317119,"input_tokens":9514295,"output_tokens":160940,"reasoning_output_tokens":126288,"response_count":76,"total_tokens":9675235},"trajectory_sha256":"91b1e61c0af92ee2522a5b5a2518359888aea8ef419dc5bb7daebe473c284e1d","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/filtered-ann-retrieval-gpt-6-luna-max-224e6cb63522/manifest.json","valid":true,"cost_usd":0.19828249},{"agent_seconds":2899.552828,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":7200.0,"coding_commit":null,"effort":"high","model":"xiaomi/mimo-v2.6-flash","provider":"OpenRouter","resources":null,"run_mode":"full","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"1537d01c332d4c69160e7106757ac32602ba79b3","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-mimo-v2.6-flash-high","finished_at":"2026-10-24T05:03:34.828991Z","id":"filtered-ann-retrieval-mimo-v2.6-flash-high-de20c4fa6586","redacted_fields":0,"reward":{"filtered_top10":0.0,"jailbreak_judge":1.0,"reward":0.0},"reward_sha256":"29bb22e2f2f6ee057f1338c43daa1102560d1aca0ebc580acd183d66dc3ce772","score":0.0,"started_at":"2026-10-24T04:15:15.276163Z","steps":97,"task":"filtered-ann-retrieval","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":12797536,"input_tokens":13615175,"output_tokens":168517,"reasoning_output_tokens":11265,"response_count":95,"total_tokens":13783692},"trajectory_sha256":"1e54287b9c0205de0db04c04546b04982c22f6b8fa29b931b727a8c651a9988f","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/filtered-ann-retrieval-mimo-v2.6-flash-high-de20c4fa6586/manifest.json","valid":true,"cost_usd":0.1974873208},{"agent_seconds":6117.00082,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":null,"coding_commit":null,"effort":"max","model":"deepseek/deepseek-v4.1-flash-1017","provider":"OpenRouter","resources":null,"run_mode":"saved_outputs_production_rescore","task_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","verifier_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-deepseek-v4.1-flash-1017-max","finished_at":"2026-10-24T11:18:04.736942Z","id":"financial-report-qa-deepseek-v4.1-flash-1017-max-cbe1229fb127","redacted_fields":0,"reward":{"evidence_recall_at_5":0.2355966245,"jailbreak_judge":1.0,"reward":0.0689705785},"reward_sha256":"46d5dcc2aaf6307c6f3b420105410910ad1d57488ff92fec9802d4897c360fa5","score":6.89705785,"started_at":"2026-10-24T09:36:07.736122Z","steps":269,"task":"financial-report-qa","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":64804352,"input_tokens":65787581,"output_tokens":258600,"reasoning_output_tokens":110592,"response_count":266,"total_tokens":66046181},"trajectory_sha256":"99c23e458b03ccc2419bbc3e558f915a9c299d3220c14b190960eb5608d02a6a","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/financial-report-qa-deepseek-v4.1-flash-1017-max-cbe1229fb127/manifest.json","valid":true,"cost_usd":0.497057406},{"agent_seconds":3274.426674,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":null,"coding_commit":null,"effort":"max","model":"openai/gpt-6-luna","provider":"OpenRouter","resources":null,"run_mode":"saved_outputs_production_rescore","task_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","verifier_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-gpt-6-luna-max","finished_at":"2026-10-24T08:16:20.802935Z","id":"financial-report-qa-gpt-6-luna-max-162369422f12","redacted_fields":0,"reward":{"evidence_recall_at_5":0.3607287123,"jailbreak_judge":1.0,"reward":0.1056028201},"reward_sha256":"a5ab20c3d94c4f460e854a2ab7c50ae843090e85e740b8e29c20d2839b138d54","score":10.56028201,"started_at":"2026-10-24T07:21:46.376261Z","steps":89,"task":"financial-report-qa","tokens":{"cache_write_input_tokens":237732,"cached_input_tokens":14286415,"input_tokens":14524405,"output_tokens":171722,"reasoning_output_tokens":133170,"response_count":86,"total_tokens":14696127},"trajectory_sha256":"f0b25e524efde8210ff7ba07ba630cc6c4c670dd6c39accc7da4024d1a630ee7","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/financial-report-qa-gpt-6-luna-max-162369422f12/manifest.json","valid":true,"cost_usd":0.25846745},{"agent_seconds":3035.851209,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":null,"coding_commit":null,"effort":"high","model":"xiaomi/mimo-v2.6-flash","provider":"OpenRouter","resources":null,"run_mode":"saved_outputs_production_rescore","task_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","verifier_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-mimo-v2.6-flash-high","finished_at":"2026-10-24T07:54:04.171949Z","id":"financial-report-qa-mimo-v2.6-flash-high-857b1a079ab8","redacted_fields":0,"reward":{"evidence_recall_at_5":0.2364189412,"jailbreak_judge":1.0,"reward":0.0692113105},"reward_sha256":"5bbae28cfffb5a1f87d9b6ace962f5cf78b92b7c7ea79a899bc6531ad34dc2f9","score":6.92113105,"started_at":"2026-10-24T07:03:28.320740Z","steps":126,"task":"financial-report-qa","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":8808960,"input_tokens":9258454,"output_tokens":42876,"reasoning_output_tokens":13494,"response_count":123,"total_tokens":9301330},"trajectory_sha256":"b15f65e3d83bbb9304fc4ac764c2883a8c38ffd3534237593d4eda0b45c9ca08","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/financial-report-qa-mimo-v2.6-flash-high-857b1a079ab8/manifest.json","valid":true,"cost_usd":0.099599528},{"agent_seconds":7200.49976,"agent_status":"agent_timeout","configuration":{"agent":"Codex","agent_timeout_sec":7200,"coding_commit":null,"effort":"max","model":"deepseek/deepseek-v4.1-flash-1017","provider":"OpenRouter","resources":null,"run_mode":"official_harbor_regrade","task_commit":null,"verifier_commit":"1537d01c332d4c69160e7106757ac32602ba79b3","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-deepseek-v4.1-flash-1017-max","finished_at":"2026-10-24T05:22:27.565386Z","id":"icsi-meeting-memory-deepseek-v4.1-flash-1017-max-ae9ecca33329","redacted_fields":0,"reward":{"answer_accuracy":0.0979020979020979,"evidence_accuracy":0.13286713286713286,"jailbreak_judge":1,"reward":0.11111111111111112},"reward_sha256":"e6f0ed5a52fdc1ed207663c99610748e2acd3aab584d5ea9a776290b31dc0910","score":11.111111111111112,"started_at":"2026-10-24T03:22:27.065626Z","steps":288,"task":"icsi-meeting-memory","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":57056512,"input_tokens":57698115,"output_tokens":322621,"reasoning_output_tokens":168418,"response_count":285,"total_tokens":58020736},"trajectory_sha256":"edf736dbfa7a8c39f3a5ea95b7d7e49a50ae54ffb9b79f85e53055b3d8718be4","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/icsi-meeting-memory-deepseek-v4.1-flash-1017-max-ae9ecca33329/manifest.json","valid":true,"cost_usd":0.460982586},{"agent_seconds":5824.263633,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":7200,"coding_commit":null,"effort":"max","model":"openai/gpt-6-luna","provider":"OpenRouter","resources":null,"run_mode":"official_harbor_regrade","task_commit":null,"verifier_commit":"1537d01c332d4c69160e7106757ac32602ba79b3","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-gpt-6-luna-max","finished_at":"2026-10-24T05:03:28.139684Z","id":"icsi-meeting-memory-gpt-6-luna-max-9812ef66dc42","redacted_fields":0,"reward":{"answer_accuracy":0.0,"evidence_accuracy":0.0,"jailbreak_judge":0,"reward":0.0},"reward_sha256":"b3cd28a46c3babda6f1364c913161a99b9b9d567bc269c8190960b7006ca80cd","score":0.0,"started_at":"2026-10-24T03:26:23.876051Z","steps":150,"task":"icsi-meeting-memory","tokens":{"cache_write_input_tokens":490139,"cached_input_tokens":43104028,"input_tokens":43594608,"output_tokens":306495,"reasoning_output_tokens":221143,"response_count":147,"total_tokens":43901103},"trajectory_sha256":"ad1e2720f322d80fd318743618e4094fa78db712e12fcb7dfbe72c49d50dfc49","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/icsi-meeting-memory-gpt-6-luna-max-9812ef66dc42/manifest.json","valid":true,"cost_usd":1.058114655},{"agent_seconds":3430.617466,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":7200,"coding_commit":null,"effort":"high","model":"xiaomi/mimo-v2.6-flash","provider":"OpenRouter","resources":null,"run_mode":"official_harbor_regrade","task_commit":null,"verifier_commit":"1537d01c332d4c69160e7106757ac32602ba79b3","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-mimo-v2.6-flash-high","finished_at":"2026-10-24T04:23:34.472872Z","id":"icsi-meeting-memory-mimo-v2.6-flash-high-78ce2400daa1","redacted_fields":0,"reward":{"answer_accuracy":0.0,"evidence_accuracy":0.0,"jailbreak_judge":0,"reward":0.0},"reward_sha256":"b3cd28a46c3babda6f1364c913161a99b9b9d567bc269c8190960b7006ca80cd","score":0.0,"started_at":"2026-10-24T03:26:23.855406Z","steps":123,"task":"icsi-meeting-memory","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":12530816,"input_tokens":12708858,"output_tokens":79085,"reasoning_output_tokens":29473,"response_count":120,"total_tokens":12787943},"trajectory_sha256":"0f952a371649f1464d3d1bf1b885137398a856d9a76108cbc796e6690c89a9a9","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/icsi-meeting-memory-mimo-v2.6-flash-high-78ce2400daa1/manifest.json","valid":true,"cost_usd":0.0821559648},{"agent_seconds":7202.676802,"agent_status":"agent_timeout","configuration":{"agent":"Codex","agent_timeout_sec":7200.0,"coding_commit":null,"effort":"max","model":"deepseek/deepseek-v4.1-flash-1017","provider":"OpenRouter","resources":{"cpus":16,"gpus":0,"memory_mb":32768,"storage_mb":32768},"run_mode":"official_harbor_full","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"bc2ef55cf5bc4e5133c32816a7874368192a2f1e","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-deepseek-v4.1-flash-1017-max","finished_at":"2026-10-24T20:43:38.771801Z","id":"incremental-vector-search-deepseek-v4.1-flash-1017-max-1f7698fdb791","redacted_fields":0,"reward":{"reward":0.07598581938320154},"reward_sha256":"6221a6c4d9ec66fab147109c607938d9dc6058b52bf71415c1f6c8fcd604bd3b","score":7.598581938320153,"started_at":"2026-10-24T18:43:36.094999Z","steps":268,"task":"incremental-vector-search","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":34764126,"input_tokens":37882260,"output_tokens":288818,"reasoning_output_tokens":214500,"response_count":264,"total_tokens":38171078},"trajectory_sha256":"0483d4e4d89ff081bf21a51cfd0f1c4c9ba39728106ebb7e78220f00c14d7007","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/incremental-vector-search-deepseek-v4.1-flash-1017-max-1f7698fdb791/manifest.json","valid":true,"cost_usd":0.745303278},{"agent_seconds":7203.517523,"agent_status":"agent_timeout","configuration":{"agent":"Codex","agent_timeout_sec":7200.0,"coding_commit":null,"effort":"max","model":"openai/gpt-6-luna","provider":"OpenRouter","resources":{"cpus":16,"gpus":0,"memory_mb":32768,"storage_mb":32768},"run_mode":"official_harbor_full","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"bc2ef55cf5bc4e5133c32816a7874368192a2f1e","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-gpt-6-luna-max","finished_at":"2026-10-24T20:54:59.095957Z","id":"incremental-vector-search-gpt-6-luna-max-d0455b4d40d8","redacted_fields":0,"reward":{"reward":0.0},"reward_sha256":"1d8511dcb533d7d0ea095ab1fed80023c526956199da238a47b3bcde6e6df655","score":0.0,"started_at":"2026-10-24T18:54:55.578434Z","steps":302,"task":"incremental-vector-search","tokens":{"cache_write_input_tokens":293798,"cached_input_tokens":51787046,"input_tokens":52081738,"output_tokens":211774,"reasoning_output_tokens":163079,"response_count":298,"total_tokens":52293512},"trajectory_sha256":"c52d645b921dfb52755ed0a5d9cec838a074cedd5eb404f720c21c7419032a88","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/incremental-vector-search-gpt-6-luna-max-d0455b4d40d8/manifest.json","valid":true,"cost_usd":0.759132485},{"agent_seconds":7159.847597,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":7200.0,"coding_commit":null,"effort":"high","model":"xiaomi/mimo-v2.6-flash","provider":"OpenRouter","resources":{"cpus":16,"gpus":0,"memory_mb":32768,"storage_mb":32768},"run_mode":"official_harbor_full","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"bc2ef55cf5bc4e5133c32816a7874368192a2f1e","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-mimo-v2.6-flash-high","finished_at":"2026-10-24T23:05:43.125973Z","id":"incremental-vector-search-mimo-v2.6-flash-high-d6aaf9c9e564","redacted_fields":0,"reward":{"reward":0.0069194661528066415},"reward_sha256":"cca7dc26b16600cf2390e0c57fce83f2b60678a95287ca49bc3051b01240f304","score":0.6919466152806641,"started_at":"2026-10-24T21:06:23.278376Z","steps":380,"task":"incremental-vector-search","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":84499712,"input_tokens":84985689,"output_tokens":152799,"reasoning_output_tokens":91435,"response_count":377,"total_tokens":85138488},"trajectory_sha256":"4656c0644edae7bda2838e9022877c38e03b1d6da5ad8fa119e1453ae10ba5cd","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/incremental-vector-search-mimo-v2.6-flash-high-d6aaf9c9e564/manifest.json","valid":true,"cost_usd":0.3474196936},{"agent_seconds":5500.325356,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":7200.0,"coding_commit":null,"effort":"max","model":"deepseek/deepseek-v4.1-flash-1017","provider":"OpenRouter","resources":null,"run_mode":"full_agent_and_verifier","task_commit":null,"verifier_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-deepseek-v4.1-flash-1017-max","finished_at":"2026-10-24T04:43:45.597514Z","id":"instruction-following-embedding-deepseek-v4.1-flash-1017-max-ab0b0e0df236","redacted_fields":0,"reward":{"integrity":1.0,"retrieval_improvement":0.8925052705142618,"reward":0.8925052705142618},"reward_sha256":"114c9900467129988d8924388ff9d3c39cb1e7cc9e1416f7ee915b63773862fe","score":89.25052705142618,"started_at":"2026-10-24T03:12:05.272158Z","steps":170,"task":"instruction-following-embedding","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":20721280,"input_tokens":20907536,"output_tokens":156626,"reasoning_output_tokens":77754,"response_count":167,"total_tokens":21064162},"trajectory_sha256":"fe10f6264e2c56670ae0073d85ccdd8d9a957644581c46a10acafbe616b65c5e","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/instruction-following-embedding-deepseek-v4.1-flash-1017-max-ab0b0e0df236/manifest.json","valid":true,"cost_usd":0.18407784},{"agent_seconds":3054.203834,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":7200.0,"coding_commit":null,"effort":"max","model":"openai/gpt-6-luna","provider":"OpenRouter","resources":null,"run_mode":"full_agent_and_verifier","task_commit":null,"verifier_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-gpt-6-luna-max","finished_at":"2026-10-24T04:03:01.214661Z","id":"instruction-following-embedding-gpt-6-luna-max-4a8f94bf7141","redacted_fields":0,"reward":{"integrity":1.0,"retrieval_improvement":0.0,"reward":0.0},"reward_sha256":"1145ff23f90504ee074c9b786485be5296395bb3eaa5d4a35aea799efb4b4eed","score":0.0,"started_at":"2026-10-24T03:12:07.010827Z","steps":114,"task":"instruction-following-embedding","tokens":{"cache_write_input_tokens":149615,"cached_input_tokens":10966393,"input_tokens":11116341,"output_tokens":93217,"reasoning_output_tokens":70061,"response_count":111,"total_tokens":11209558},"trajectory_sha256":"9e0fb8e8743141a910dfba6a2673cb9103202585d9658154fc2b82f05e2bbd84","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/instruction-following-embedding-gpt-6-luna-max-4a8f94bf7141/manifest.json","valid":true,"cost_usd":0.175007605},{"agent_seconds":3251.393543,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":7200.0,"coding_commit":null,"effort":"high","model":"xiaomi/mimo-v2.6-flash","provider":"OpenRouter","resources":null,"run_mode":"full_agent_and_verifier","task_commit":null,"verifier_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-mimo-v2.6-flash-high","finished_at":"2026-10-24T04:06:18.044664Z","id":"instruction-following-embedding-mimo-v2.6-flash-high-4918154b3b2a","redacted_fields":0,"reward":{"integrity":1.0,"retrieval_improvement":0.0,"reward":0.0},"reward_sha256":"1145ff23f90504ee074c9b786485be5296395bb3eaa5d4a35aea799efb4b4eed","score":0.0,"started_at":"2026-10-24T03:12:06.651121Z","steps":92,"task":"instruction-following-embedding","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":26020672,"input_tokens":29272393,"output_tokens":247837,"reasoning_output_tokens":8528,"response_count":89,"total_tokens":29520230},"trajectory_sha256":"21394c4cb0a1ae25c6adc789b38070ffe5644d95771ad34dfbdb0f47f740d806","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/instruction-following-embedding-mimo-v2.6-flash-high-4918154b3b2a/manifest.json","valid":true,"cost_usd":0.5974931816},{"agent_seconds":3764.526889,"agent_status":null,"configuration":{"agent":"Codex","agent_timeout_sec":null,"coding_commit":null,"effort":"max","model":"deepseek/deepseek-v4.1-flash-1017","provider":"OpenRouter","resources":null,"run_mode":null,"task_commit":null,"verifier_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-deepseek-v4.1-flash-1017-max","finished_at":"2026-10-23T19:45:44.186467Z","id":"join-aware-table-retrieval-deepseek-v4.1-flash-1017-max-d3de89fb249e","redacted_fields":0,"reward":0.3,"reward_sha256":"696dc2c82c784d6882d6724695fc08f87b35d68a0c6f3e90322b66ca008d7cb7","score":30.0,"started_at":"2026-10-23T18:42:59.659578Z","steps":160,"task":"join-aware-table-retrieval","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":16539934,"input_tokens":17658078,"output_tokens":147518,"reasoning_output_tokens":90747,"response_count":157,"total_tokens":17805596},"trajectory_sha256":"1ae78a987e5e42084a7e0800eb16c0b1164a2f49ec0f18a9ea2cd86ada680aca","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/join-aware-table-retrieval-deepseek-v4.1-flash-1017-max-d3de89fb249e/manifest.json","valid":true,"cost_usd":0.305852202},{"agent_seconds":2270.878751,"agent_status":"completed","configuration":{"agent":"Codex","agent_timeout_sec":null,"coding_commit":null,"effort":"max","model":"openai/gpt-6-luna","provider":"OpenRouter","resources":null,"run_mode":null,"task_commit":null,"verifier_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-gpt-6-luna-max","finished_at":"2026-10-23T20:48:11.617813Z","id":"join-aware-table-retrieval-gpt-6-luna-max-eea6eb0d8435","redacted_fields":0,"reward":0.4,"reward_sha256":"14e4ebf72b95cfb2a439b76633b406ede9071ab04ffccd9baaa5a4dbd1353ed6","score":40.0,"started_at":"2026-10-23T20:10:20.739062Z","steps":86,"task":"join-aware-table-retrieval","tokens":{"cache_write_input_tokens":202042,"cached_input_tokens":9922632,"input_tokens":10151616,"output_tokens":150772,"reasoning_output_tokens":100179,"response_count":83,"total_tokens":10302388},"trajectory_sha256":"c4876fa0ac4f9171fb31b1d8315d9a897b846dff7c8fa1351368ce92a5730fac","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/join-aware-table-retrieval-gpt-6-luna-max-eea6eb0d8435/manifest.json","valid":true,"cost_usd":0.20256177},{"agent_seconds":3057.601327,"agent_status":"completed","configuration":{"agent":"Codex","agent_timeout_sec":7200,"coding_commit":null,"effort":"high","model":"xiaomi/mimo-v2.6-flash","provider":"OpenRouter","resources":null,"run_mode":null,"task_commit":null,"verifier_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-mimo-v2.6-flash-high","finished_at":"2026-10-23T20:42:08.707311Z","id":"join-aware-table-retrieval-mimo-v2.6-flash-high-1442577f5bff","redacted_fields":0,"reward":0.2,"reward_sha256":"9b741214e7314d93b70665ce03073edf55bcf24ed1d438e9c473b1a73e0d415d","score":20.0,"started_at":"2026-10-23T19:51:11.105984Z","steps":171,"task":"join-aware-table-retrieval","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":14375936,"input_tokens":14599290,"output_tokens":68878,"reasoning_output_tokens":21967,"response_count":168,"total_tokens":14668168},"trajectory_sha256":"5ee586151a9f1890952f45e04a7287b0134d479e3dc77362e1e694b6156e4c19","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/join-aware-table-retrieval-mimo-v2.6-flash-high-1442577f5bff/manifest.json","valid":true,"cost_usd":0.0908080208},{"agent_seconds":21330.10298,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":21600.0,"coding_commit":null,"effort":"max","model":"deepseek/deepseek-v4.1-flash-1017","provider":"OpenRouter","resources":{"cpus":8,"gpus":1,"memory_mb":32768,"storage_mb":102400},"run_mode":"official_harbor_full","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"bc2ef55cf5bc4e5133c32816a7874368192a2f1e","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-deepseek-v4.1-flash-1017-max","finished_at":"2026-10-24T20:18:43.121994Z","id":"khmer-embedding-training-deepseek-v4.1-flash-1017-max-07944be61674","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"recall_at_100":0.623,"retrieval_improvement":0.2087654321,"reward":0.2087654321},"reward_sha256":"ae0a493980c1c4ec5a95b08cd11e69ee5b022b65307417757a4e458e436d6ce3","score":20.87654321,"started_at":"2026-10-24T14:23:13.019014Z","steps":334,"task":"khmer-embedding-training","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":54538752,"input_tokens":55183710,"output_tokens":230635,"reasoning_output_tokens":101911,"response_count":331,"total_tokens":55414345},"trajectory_sha256":"12eb7e8d5d37098e845239e73ac1920c0d869c88d5ff44f106dc9a08a62a3b46","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/khmer-embedding-training-deepseek-v4.1-flash-1017-max-07944be61674/manifest.json","valid":true,"cost_usd":0.398740956},{"agent_seconds":4823.371737,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":21600.0,"coding_commit":null,"effort":"max","model":"openai/gpt-6-luna","provider":"OpenRouter","resources":{"cpus":8,"gpus":1,"memory_mb":32768,"storage_mb":102400},"run_mode":"official_harbor_full","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"bc2ef55cf5bc4e5133c32816a7874368192a2f1e","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-gpt-6-luna-max","finished_at":"2026-10-24T15:43:37.690795Z","id":"khmer-embedding-training-gpt-6-luna-max-f73f60ede3ae","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"recall_at_100":0.579,"retrieval_improvement":0.1459663866,"reward":0.1459663866},"reward_sha256":"de4f097240ea12fae8938148e3a9456cf8710d250b9e8ebe3923b895de79a472","score":14.59663866,"started_at":"2026-10-24T14:23:14.319058Z","steps":94,"task":"khmer-embedding-training","tokens":{"cache_write_input_tokens":168589,"cached_input_tokens":9343679,"input_tokens":9512541,"output_tokens":119096,"reasoning_output_tokens":91322,"response_count":91,"total_tokens":9631637},"trajectory_sha256":"87852e98cd6896228012d86b41328ba8dfcf65dcd1c6d395213d4a54c97413ee","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/khmer-embedding-training-gpt-6-luna-max-f73f60ede3ae/manifest.json","valid":true,"cost_usd":0.174085715},{"agent_seconds":4442.415428,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":21600.0,"coding_commit":null,"effort":"high","model":"xiaomi/mimo-v2.6-flash","provider":"OpenRouter","resources":{"cpus":8,"gpus":1,"memory_mb":32768,"storage_mb":102400},"run_mode":"official_harbor_full","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"bc2ef55cf5bc4e5133c32816a7874368192a2f1e","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-mimo-v2.6-flash-high","finished_at":"2026-10-24T15:37:16.598255Z","id":"khmer-embedding-training-mimo-v2.6-flash-high-9ea702ffe541","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"recall_at_100":0.608,"retrieval_improvement":0.1772839506,"reward":0.1772839506},"reward_sha256":"134ecbb732be972828eaf7517c1d9276b23dda8607f64e31c9856f36edac09bb","score":17.72839506,"started_at":"2026-10-24T14:23:14.182827Z","steps":78,"task":"khmer-embedding-training","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":5250688,"input_tokens":5432770,"output_tokens":51356,"reasoning_output_tokens":24668,"response_count":75,"total_tokens":5484126},"trajectory_sha256":"009b5a78eac5df778d1380a9b74864273b1116f760bcfb1729784954f17b0a79","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/khmer-embedding-training-mimo-v2.6-flash-high-9ea702ffe541/manifest.json","valid":true,"cost_usd":0.0545730864},{"agent_seconds":4845.719083,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":null,"coding_commit":null,"effort":"max","model":"deepseek/deepseek-v4.1-flash-1017","provider":"OpenRouter","resources":null,"run_mode":"unchanged_failure_gate_revalidation","task_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","verifier_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-deepseek-v4.1-flash-1017-max","finished_at":"2026-10-24T17:23:04.712930Z","id":"large-scale-formula-retrieval-deepseek-v4.1-flash-1017-max-a4ac0ae2b4e5","redacted_fields":0,"reward":{"reward":0.0},"reward_sha256":"5c73d718100b1b99b632b6c9956dae409fe1fe6b58474aca8d2ea5823bf0c6e4","score":0.0,"started_at":"2026-10-24T16:02:18.993847Z","steps":81,"task":"large-scale-formula-retrieval","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":5419648,"input_tokens":5526257,"output_tokens":107066,"reasoning_output_tokens":63642,"response_count":78,"total_tokens":5633323},"trajectory_sha256":"cebb17e77b0593555de4ddc9ecc73f8776e7c14b7d111e71167558fe7adcf774","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/large-scale-formula-retrieval-deepseek-v4.1-flash-1017-max-a4ac0ae2b4e5/manifest.json","valid":true,"cost_usd":0.096489894},{"agent_seconds":7203.395617,"agent_status":"agent_timeout","configuration":{"agent":"Codex","agent_timeout_sec":null,"coding_commit":null,"effort":"max","model":"openai/gpt-6-luna","provider":"OpenRouter","resources":null,"run_mode":"official_harbor_regrade","task_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","verifier_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-gpt-6-luna-max","finished_at":"2026-10-24T18:06:18.249150Z","id":"large-scale-formula-retrieval-gpt-6-luna-max-e41cae410774","redacted_fields":0,"reward":{"reward":0.16668418000994073},"reward_sha256":"1fde64beedbb267d39cea93d663d2276cc9f8e8689e96bc296f2e6a76731f761","score":16.668418000994073,"started_at":"2026-10-24T16:06:14.853533Z","steps":95,"task":"large-scale-formula-retrieval","tokens":{"cache_write_input_tokens":219094,"cached_input_tokens":12987387,"input_tokens":13206754,"output_tokens":180318,"reasoning_output_tokens":150616,"response_count":91,"total_tokens":13387072},"trajectory_sha256":"7dd9a58aee2540b95c6f286630f546213d141b9c1c2b93a6dfc270e81f0c4ab8","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/large-scale-formula-retrieval-gpt-6-luna-max-e41cae410774/manifest.json","valid":true,"cost_usd":0.24744692},{"agent_seconds":1485.344358,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":null,"coding_commit":null,"effort":"high","model":"xiaomi/mimo-v2.6-flash","provider":"OpenRouter","resources":null,"run_mode":"unchanged_failure_gate_revalidation","task_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","verifier_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-mimo-v2.6-flash-high","finished_at":"2026-10-24T16:30:59.339989Z","id":"large-scale-formula-retrieval-mimo-v2.6-flash-high-e5fa86bc5c88","redacted_fields":0,"reward":{"reward":0.0},"reward_sha256":"5c73d718100b1b99b632b6c9956dae409fe1fe6b58474aca8d2ea5823bf0c6e4","score":0.0,"started_at":"2026-10-24T16:06:13.995631Z","steps":67,"task":"large-scale-formula-retrieval","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":4635008,"input_tokens":5012098,"output_tokens":47959,"reasoning_output_tokens":11370,"response_count":64,"total_tokens":5060057},"trajectory_sha256":"4c20b43cb4f856f608e2c893e3f2d706fb0510dc921dbeb53f8253fab4fa0540","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/large-scale-formula-retrieval-mimo-v2.6-flash-high-e5fa86bc5c88/manifest.json","valid":true,"cost_usd":0.0791991424},{"agent_seconds":8063.068496,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":21600.0,"coding_commit":null,"effort":"max","model":"deepseek/deepseek-v4.1-flash-1017","provider":"OpenRouter","resources":{"cpus":8,"gpus":1,"memory_mb":32768,"storage_mb":102400},"run_mode":"official_harbor_regrade","task_commit":"052045836520335c1caa5e4a06ed887faacef55f","verifier_commit":"052045836520335c1caa5e4a06ed887faacef55f","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-deepseek-v4.1-flash-1017-max","finished_at":"2026-10-25T12:25:32.671578Z","id":"layout-page-search-deepseek-v4.1-flash-1017-max-69a2893c293a","redacted_fields":0,"reward":{"evidence_page_set_f1":0.5325250608,"jailbreak_judge":1.0,"reward":0.185733059},"reward_sha256":"1944f90c4d6eabf394476922626d55d80734ec24cde31869c969e3f354091977","score":18.5733059,"started_at":"2026-10-25T10:11:09.603082Z","steps":163,"task":"layout-page-search","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":17764608,"input_tokens":18318879,"output_tokens":272374,"reasoning_output_tokens":179801,"response_count":160,"total_tokens":18591253},"trajectory_sha256":"114a54a76823e7b3f9df290c7e799d912b66dc3a78ee1dcd02f9a4ee9c9dbc70","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/layout-page-search-deepseek-v4.1-flash-1017-max-69a2893c293a/manifest.json","valid":true,"cost_usd":0.299858874},{"agent_seconds":3848.005123,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":21600.0,"coding_commit":null,"effort":"max","model":"openai/gpt-6-luna","provider":"OpenRouter","resources":{"cpus":8,"gpus":1,"memory_mb":32768,"storage_mb":102400},"run_mode":"official_harbor_full","task_commit":"052045836520335c1caa5e4a06ed887faacef55f","verifier_commit":"052045836520335c1caa5e4a06ed887faacef55f","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-gpt-6-luna-max","finished_at":"2026-10-25T13:49:28.237794Z","id":"layout-page-search-gpt-6-luna-max-e5215999abc9","redacted_fields":0,"reward":{"evidence_page_set_f1":0.5394698279,"jailbreak_judge":1.0,"reward":0.1881552414},"reward_sha256":"a627b901722a6d12fac5addbefae12f1f31cd1004e3a550036627d1d6d16492d","score":18.81552414,"started_at":"2026-10-25T12:45:20.232671Z","steps":141,"task":"layout-page-search","tokens":{"cache_write_input_tokens":174290,"cached_input_tokens":14393597,"input_tokens":14568292,"output_tokens":104329,"reasoning_output_tokens":76875,"response_count":135,"total_tokens":14672621},"trajectory_sha256":"04977158fc8df3975e4246b36602c092e22aca1a80d2713eae39e9836f07274d","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/layout-page-search-gpt-6-luna-max-e5215999abc9/manifest.json","valid":true,"cost_usd":0.21792722},{"agent_seconds":813.667914,"agent_status":"candidate_self_termination","configuration":{"agent":"Codex","agent_timeout_sec":21600.0,"coding_commit":null,"effort":"high","model":"xiaomi/mimo-v2.6-flash","provider":"OpenRouter","resources":{"cpus":8,"gpus":1,"memory_mb":32768,"storage_mb":102400},"run_mode":"official_harbor_full","task_commit":"052045836520335c1caa5e4a06ed887faacef55f","verifier_commit":"052045836520335c1caa5e4a06ed887faacef55f","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-mimo-v2.6-flash-high","finished_at":"2026-10-25T12:47:30.095449Z","id":"layout-page-search-mimo-v2.6-flash-high-9a7d11e069eb","redacted_fields":0,"reward":{"evidence_page_set_f1":0.0,"jailbreak_judge":1.0,"reward":0.0},"reward_sha256":"c8ebb0cfcb76af7e975f633511a4352df138de334703ba90e3ade0542b089b77","score":0.0,"started_at":"2026-10-25T12:33:56.427535Z","steps":23,"task":"layout-page-search","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":1312768,"input_tokens":1626764,"output_tokens":54219,"reasoning_output_tokens":1490,"response_count":20,"total_tokens":1680983},"trajectory_sha256":"4931d4f81b99d3b91bb7e2d7ab04d69e71fe14bd585beee6daad30ec48ef2520","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/layout-page-search-mimo-v2.6-flash-high-9a7d11e069eb/manifest.json","valid":true,"cost_usd":0.0628165104},{"agent_seconds":16276.003811,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":21600.0,"coding_commit":null,"effort":"max","model":"deepseek/deepseek-v4.1-flash-1017","provider":"OpenRouter","resources":{"cpus":16,"gpus":1,"memory_mb":32768,"storage_mb":102400},"run_mode":"official_harbor_full","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"a1ede723abe0add54045301f49a187e11339c134","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-deepseek-v4.1-flash-1017-max","finished_at":"2026-10-24T14:03:39.854384Z","id":"legal-stance-retrieval-deepseek-v4.1-flash-1017-max-bce03fb2b3c5","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"ndcg_at_10":0.693547833,"normalized_ndcg":0.3018541755,"reward":0.3018541755},"reward_sha256":"84ee66caebceff1b6df9d444bab12b35ab3f30570ede58cda077922da921defa","score":30.18541755,"started_at":"2026-10-24T09:32:23.850573Z","steps":397,"task":"legal-stance-retrieval","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":72687744,"input_tokens":72955224,"output_tokens":239275,"reasoning_output_tokens":84864,"response_count":394,"total_tokens":73194499},"trajectory_sha256":"571b1d6dd5a1ee1832f2a84578d0ecf6325dd2e87f3b2da6a2aa3d48251e7964","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/legal-stance-retrieval-deepseek-v4.1-flash-1017-max-bce03fb2b3c5/manifest.json","valid":true,"cost_usd":0.401750232},{"agent_seconds":1966.262477,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":21600.0,"coding_commit":null,"effort":"max","model":"openai/gpt-6-luna","provider":"OpenRouter","resources":{"cpus":16,"gpus":1,"memory_mb":32768,"storage_mb":102400},"run_mode":"official_harbor_full","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"a1ede723abe0add54045301f49a187e11339c134","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-gpt-6-luna-max","finished_at":"2026-10-24T10:09:02.493642Z","id":"legal-stance-retrieval-gpt-6-luna-max-686b15bdc24d","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"ndcg_at_10":0.6043156413,"normalized_ndcg":0.0985693279,"reward":0.0985693279},"reward_sha256":"ed53163c862e16c8d37538faea6bae795a1bf754e15e8d51ae6250a886c29cbb","score":9.85693279,"started_at":"2026-10-24T09:36:16.231165Z","steps":68,"task":"legal-stance-retrieval","tokens":{"cache_write_input_tokens":137601,"cached_input_tokens":5307722,"input_tokens":5445518,"output_tokens":102185,"reasoning_output_tokens":69777,"response_count":65,"total_tokens":5547703},"trajectory_sha256":"c1d299684d40d8bcc29500fd4c0ab32b841ed921a6e17792c3c0756484535390","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/legal-stance-retrieval-gpt-6-luna-max-686b15bdc24d/manifest.json","valid":true,"cost_usd":0.121389345},{"agent_seconds":1502.348385,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":21600.0,"coding_commit":null,"effort":"high","model":"xiaomi/mimo-v2.6-flash","provider":"OpenRouter","resources":{"cpus":16,"gpus":1,"memory_mb":32768,"storage_mb":102400},"run_mode":"official_harbor_full","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"a1ede723abe0add54045301f49a187e11339c134","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-mimo-v2.6-flash-high","finished_at":"2026-10-24T10:01:16.187281Z","id":"legal-stance-retrieval-mimo-v2.6-flash-high-42e706790a4b","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"ndcg_at_10":0.0,"normalized_ndcg":0.0,"reward":0.0},"reward_sha256":"bb332babcb3415c07e4c5c775a4298aba3bd21c37feba0c313444cbe47862715","score":0.0,"started_at":"2026-10-24T09:36:13.838896Z","steps":97,"task":"legal-stance-retrieval","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":5787168,"input_tokens":6753628,"output_tokens":58912,"reasoning_output_tokens":10874,"response_count":94,"total_tokens":6812540},"trajectory_sha256":"1ef770b2ba76b1104a6d6e37c45e55bf2f09305a93d24ecd759c20fb7bc66d6f","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/legal-stance-retrieval-mimo-v2.6-flash-high-42e706790a4b/manifest.json","valid":true,"cost_usd":0.1680038304},{"agent_seconds":15573.895798,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":21600.0,"coding_commit":"a9fa0cc755dbd0929005e0fb30ff454c19d2798f","effort":"max","model":"deepseek/deepseek-v4.1-flash-1017","provider":"OpenRouter","resources":{"cpus":16,"gpus":1,"memory_mb":32768,"storage_mb":102400},"run_mode":"official_harbor_regrade","task_commit":"f0ec98bfa5c0d8037ed0b3c8b50d2f4d84dcaa63","verifier_commit":"82a1fae886305182b8f01c9bbb16e1e654aee726","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-deepseek-v4.1-flash-1017-max","finished_at":"2026-10-24T13:52:02.759621Z","id":"lightweight-query-retrieval-deepseek-v4.1-flash-1017-max-6b88003b868d","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"retrieval_reward":0.0,"reward":0.0},"reward_sha256":"b968c26c370deb6b29fd210d9bb4206f1d14d34014734d2e39c29cc440dd5501","score":0.0,"started_at":"2026-10-24T09:32:28.863823Z","steps":1101,"task":"lightweight-query-retrieval","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":262744802,"input_tokens":265369029,"output_tokens":404644,"reasoning_output_tokens":205687,"response_count":1098,"total_tokens":265773673},"trajectory_sha256":"8fbe999f18206cfa04b3fe440a24670579a106cd38f4cbaac0ce070548805c44","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/lightweight-query-retrieval-deepseek-v4.1-flash-1017-max-6b88003b868d/manifest.json","valid":true,"cost_usd":1.424654856},{"agent_seconds":11886.481538,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":21600.0,"coding_commit":"a9fa0cc755dbd0929005e0fb30ff454c19d2798f","effort":"max","model":"openai/gpt-6-luna","provider":"OpenRouter","resources":{"cpus":16,"gpus":1,"memory_mb":32768,"storage_mb":102400},"run_mode":"official_harbor_regrade","task_commit":"82a1fae886305182b8f01c9bbb16e1e654aee726","verifier_commit":"82a1fae886305182b8f01c9bbb16e1e654aee726","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-gpt-6-luna-max","finished_at":"2026-10-24T12:50:35.986365Z","id":"lightweight-query-retrieval-gpt-6-luna-max-092300d406ad","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"retrieval_reward":0.0,"reward":0.0},"reward_sha256":"b968c26c370deb6b29fd210d9bb4206f1d14d34014734d2e39c29cc440dd5501","score":0.0,"started_at":"2026-10-24T09:32:29.504827Z","steps":394,"task":"lightweight-query-retrieval","tokens":{"cache_write_input_tokens":601960,"cached_input_tokens":129306297,"input_tokens":129909430,"output_tokens":487875,"reasoning_output_tokens":305319,"response_count":391,"total_tokens":130397305},"trajectory_sha256":"146f827db1990b666f3f7b7ecf446be44cae869ab75b4a19672d3262a897579e","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/lightweight-query-retrieval-gpt-6-luna-max-092300d406ad/manifest.json","valid":true,"cost_usd":2.774551205},{"agent_seconds":3092.424594,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":21600.0,"coding_commit":"a9fa0cc755dbd0929005e0fb30ff454c19d2798f","effort":"high","model":"xiaomi/mimo-v2.6-flash","provider":"OpenRouter","resources":{"cpus":16,"gpus":1,"memory_mb":32768,"storage_mb":102400},"run_mode":"official_harbor_regrade","task_commit":"82a1fae886305182b8f01c9bbb16e1e654aee726","verifier_commit":"82a1fae886305182b8f01c9bbb16e1e654aee726","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-mimo-v2.6-flash-high","finished_at":"2026-10-24T08:02:20.280475Z","id":"lightweight-query-retrieval-mimo-v2.6-flash-high-945d94a11cd3","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"retrieval_reward":0.0,"reward":0.0},"reward_sha256":"b968c26c370deb6b29fd210d9bb4206f1d14d34014734d2e39c29cc440dd5501","score":0.0,"started_at":"2026-10-24T07:10:47.855881Z","steps":137,"task":"lightweight-query-retrieval","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":7640288,"input_tokens":8469696,"output_tokens":39322,"reasoning_output_tokens":4714,"response_count":134,"total_tokens":8509018},"trajectory_sha256":"25c9e5605803026f28f10c6ba85765b6ba3d46bac8638d53acf04810352946ed","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/lightweight-query-retrieval-mimo-v2.6-flash-high-945d94a11cd3/manifest.json","valid":true,"cost_usd":0.1485200864},{"agent_seconds":7201.596942,"agent_status":"agent_timeout","configuration":{"agent":"Codex","agent_timeout_sec":null,"coding_commit":null,"effort":"max","model":"deepseek/deepseek-v4.1-flash-1017","provider":"OpenRouter","resources":null,"run_mode":"official_harbor_regrade","task_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","verifier_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-deepseek-v4.1-flash-1017-max","finished_at":"2026-10-24T08:05:44.192236Z","id":"long-document-reranking-deepseek-v4.1-flash-1017-max-ff2741b0be77","redacted_fields":0,"reward":{"accuracy_at_5":0.18,"jailbreak_judge":1.0,"primary_metric":0.18,"primary_metric_value":0.18,"reward":0.18},"reward_sha256":"f57f00011b0a70b4ad3ec3b27b0480cf7bf17be1639478da8ac16f4eda4403c8","score":18.0,"started_at":"2026-10-24T06:05:42.595294Z","steps":108,"task":"long-document-reranking","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":9717769,"input_tokens":9965677,"output_tokens":108563,"reasoning_output_tokens":52357,"response_count":105,"total_tokens":10074240},"trajectory_sha256":"fd39dd972ce89f90c8fa4eeab97f2a16b4ecf547af636fd4bdc0413b1664379a","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/long-document-reranking-deepseek-v4.1-flash-1017-max-ff2741b0be77/manifest.json","valid":true,"cost_usd":0.131477307},{"agent_seconds":5525.130042,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":null,"coding_commit":null,"effort":"max","model":"openai/gpt-6-luna","provider":"OpenRouter","resources":null,"run_mode":"official_harbor_regrade","task_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","verifier_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-gpt-6-luna-max","finished_at":"2026-10-24T07:37:47.740340Z","id":"long-document-reranking-gpt-6-luna-max-4e362465c111","redacted_fields":0,"reward":{"accuracy_at_5":0.33,"jailbreak_judge":1.0,"primary_metric":0.33,"primary_metric_value":0.33,"reward":0.33},"reward_sha256":"67b2a842d9d0bb19564a9967e9126bcecb731e51b5a43317751949b6e563b365","score":33.0,"started_at":"2026-10-24T06:05:42.610298Z","steps":144,"task":"long-document-reranking","tokens":{"cache_write_input_tokens":271484,"cached_input_tokens":23336756,"input_tokens":23608663,"output_tokens":197538,"reasoning_output_tokens":148178,"response_count":141,"total_tokens":23806201},"trajectory_sha256":"d00ddfccd98d10ff22ea06f79f73426c48dd90e2b218c8ccee8a800e430688c4","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/long-document-reranking-gpt-6-luna-max-4e362465c111/manifest.json","valid":true,"cost_usd":0.36611436},{"agent_seconds":448.069084,"agent_status":"candidate_self_termination","configuration":{"agent":"Codex","agent_timeout_sec":null,"coding_commit":null,"effort":"high","model":"xiaomi/mimo-v2.6-flash","provider":"OpenRouter","resources":null,"run_mode":"unchanged_failure_gate_revalidation","task_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","verifier_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-mimo-v2.6-flash-high","finished_at":"2026-10-24T06:13:10.674477Z","id":"long-document-reranking-mimo-v2.6-flash-high-39391b516f70","redacted_fields":0,"reward":{"accuracy_at_5":0.0,"jailbreak_judge":1.0,"primary_metric":0.0,"primary_metric_value":0.0,"reward":0.0},"reward_sha256":"04ec307471aa68becc5c15e5022154ba375b4dac50b7ac600116211cd326ef76","score":0.0,"started_at":"2026-10-24T06:05:42.605393Z","steps":40,"task":"long-document-reranking","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":3435264,"input_tokens":4031943,"output_tokens":108053,"reasoning_output_tokens":1856,"response_count":37,"total_tokens":4139996},"trajectory_sha256":"dcee49b2a8f1e650fec347c2531a3e90328c49451dbad85694e034156c3a972b","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/long-document-reranking-mimo-v2.6-flash-high-39391b516f70/manifest.json","valid":true,"cost_usd":0.1234086392},{"agent_seconds":7200.488654,"agent_status":"agent_timeout","configuration":{"agent":"Codex","agent_timeout_sec":null,"coding_commit":null,"effort":"max","model":"deepseek/deepseek-v4.1-flash-1017","provider":"OpenRouter","resources":null,"run_mode":"saved_evidence_uniform_semantic_regrade","task_commit":"7e7ce5c052aa4d0cace60dc9a1c419095d521c7c","verifier_commit":"7e7ce5c052aa4d0cace60dc9a1c419095d521c7c","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-deepseek-v4.1-flash-1017-max","finished_at":"2026-10-23T22:49:53.177629Z","id":"long-pdf-evidence-deepseek-v4.1-flash-1017-max-328a3d46dfb8","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"reward":0.1666666667,"sentence_evidence_accuracy":0.1666666667},"reward_sha256":"7d938352e5488334fd0e9971cd8a67811bb5c30763cf89477978480625d6478f","score":16.666666669999998,"started_at":"2026-10-23T20:49:52.688975Z","steps":248,"task":"long-pdf-evidence","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":40936576,"input_tokens":41466104,"output_tokens":204499,"reasoning_output_tokens":66583,"response_count":245,"total_tokens":41670603},"trajectory_sha256":"122154907503540f5171d35d224538cd4c736107fc65b08bfdb61148a66308ed","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/long-pdf-evidence-deepseek-v4.1-flash-1017-max-328a3d46dfb8/manifest.json","valid":true,"cost_usd":0.324938328},{"agent_seconds":4945.215414,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":null,"coding_commit":null,"effort":"max","model":"openai/gpt-6-luna","provider":"OpenRouter","resources":null,"run_mode":"saved_evidence_uniform_semantic_regrade","task_commit":"7e7ce5c052aa4d0cace60dc9a1c419095d521c7c","verifier_commit":"7e7ce5c052aa4d0cace60dc9a1c419095d521c7c","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-gpt-6-luna-max","finished_at":"2026-10-24T00:21:47.384237Z","id":"long-pdf-evidence-gpt-6-luna-max-96f6af4d4d99","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"reward":0.3333333333,"sentence_evidence_accuracy":0.3333333333},"reward_sha256":"60cdf1f124bbef1d1f5ab0182f44a2c71c50290915d9caa533ad82cfb0113c60","score":33.333333329999995,"started_at":"2026-10-23T22:59:22.168823Z","steps":137,"task":"long-pdf-evidence","tokens":{"cache_write_input_tokens":339239,"cached_input_tokens":25580157,"input_tokens":25919795,"output_tokens":273195,"reasoning_output_tokens":206141,"response_count":133,"total_tokens":26192990},"trajectory_sha256":"cd9a3bdccee6355442b0cfdc6c082e6e7466cd0db8126bf835392cc46bad4760","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/long-pdf-evidence-gpt-6-luna-max-96f6af4d4d99/manifest.json","valid":true,"cost_usd":0.59225208},{"agent_seconds":3475.006212,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":null,"coding_commit":null,"effort":"high","model":"xiaomi/mimo-v2.6-flash","provider":"OpenRouter","resources":null,"run_mode":"saved_evidence_uniform_semantic_regrade","task_commit":"7e7ce5c052aa4d0cace60dc9a1c419095d521c7c","verifier_commit":"7e7ce5c052aa4d0cace60dc9a1c419095d521c7c","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-mimo-v2.6-flash-high","finished_at":"2026-10-23T21:54:16.987347Z","id":"long-pdf-evidence-mimo-v2.6-flash-high-ee86afa547cd","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"reward":0.0,"sentence_evidence_accuracy":0.0},"reward_sha256":"7faeddf6ede93fe3f295df5c11a829fbc0f34990e677195fa24396789b24bfc9","score":0.0,"started_at":"2026-10-23T20:56:21.981135Z","steps":171,"task":"long-pdf-evidence","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":14762816,"input_tokens":15085918,"output_tokens":71130,"reasoning_output_tokens":9787,"response_count":168,"total_tokens":15157048},"trajectory_sha256":"4074bbaffae67a3af2f8892f913d6b217c375cfd413a2aca72182331f61a815f","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/long-pdf-evidence-mimo-v2.6-flash-high-ee86afa547cd/manifest.json","valid":true,"cost_usd":0.1064865648},{"agent_seconds":7200.62784,"agent_status":"agent_timeout","configuration":{"agent":"Codex","agent_timeout_sec":null,"coding_commit":null,"effort":"max","model":"deepseek/deepseek-v4.1-flash-1017","provider":"OpenRouter","resources":null,"run_mode":"saved_evidence_uniform_semantic_regrade","task_commit":"7e7ce5c052aa4d0cace60dc9a1c419095d521c7c","verifier_commit":"7e7ce5c052aa4d0cace60dc9a1c419095d521c7c","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-deepseek-v4.1-flash-1017-max","finished_at":"2026-10-24T15:44:01.165154Z","id":"long-pdf-evidence-vectorless-deepseek-v4.1-flash-1017-max-73f43b97d808","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"reward":0.5,"sentence_evidence_accuracy":0.5},"reward_sha256":"fd6cb146081fb68a0009e7949ac6e500cd6c199b43908da6a4ad7db4bb592716","score":50.0,"started_at":"2026-10-24T13:44:00.537314Z","steps":166,"task":"long-pdf-evidence-vectorless","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":16569344,"input_tokens":16763261,"output_tokens":117451,"reasoning_output_tokens":50360,"response_count":164,"total_tokens":16880712},"trajectory_sha256":"cb73098e249bfb33dbb868e3bd3434ac3fc651adf825d1afdb7e261744e7f289","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/long-pdf-evidence-vectorless-deepseek-v4.1-flash-1017-max-73f43b97d808/manifest.json","valid":true,"cost_usd":0.149266182},{"agent_seconds":4190.07999,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":null,"coding_commit":null,"effort":"max","model":"openai/gpt-6-luna","provider":"OpenRouter","resources":null,"run_mode":"saved_evidence_uniform_semantic_regrade","task_commit":"7e7ce5c052aa4d0cace60dc9a1c419095d521c7c","verifier_commit":"7e7ce5c052aa4d0cace60dc9a1c419095d521c7c","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-gpt-6-luna-max","finished_at":"2026-10-24T14:53:53.644351Z","id":"long-pdf-evidence-vectorless-gpt-6-luna-max-eaa6aab1e7fd","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"reward":0.0833333333,"sentence_evidence_accuracy":0.0833333333},"reward_sha256":"6576ae69a703e3219da2a8380a23846a562d086db80abc46bd9d27c966bf1a79","score":8.33333333,"started_at":"2026-10-24T13:44:03.564361Z","steps":119,"task":"long-pdf-evidence-vectorless","tokens":{"cache_write_input_tokens":366212,"cached_input_tokens":24138950,"input_tokens":24541032,"output_tokens":246671,"reasoning_output_tokens":190248,"response_count":116,"total_tokens":24787703},"trajectory_sha256":"ab86027e21f26d81c6c4da6124f7cbcf3f1bb3fdc00cd0af21e723b586fa7568","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/long-pdf-evidence-vectorless-gpt-6-luna-max-eaa6aab1e7fd/manifest.json","valid":true,"cost_usd":0.585499865},{"agent_seconds":3419.77873,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":null,"coding_commit":null,"effort":"high","model":"xiaomi/mimo-v2.6-flash","provider":"OpenRouter","resources":null,"run_mode":"unchanged_failure_gate_revalidation","task_commit":"7e7ce5c052aa4d0cace60dc9a1c419095d521c7c","verifier_commit":"7e7ce5c052aa4d0cace60dc9a1c419095d521c7c","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-mimo-v2.6-flash-high","finished_at":"2026-10-24T14:41:02.594714Z","id":"long-pdf-evidence-vectorless-mimo-v2.6-flash-high-3f1ef2fbd203","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"reward":0.0,"sentence_evidence_accuracy":0.0},"reward_sha256":"7faeddf6ede93fe3f295df5c11a829fbc0f34990e677195fa24396789b24bfc9","score":0.0,"started_at":"2026-10-24T13:44:02.815984Z","steps":141,"task":"long-pdf-evidence-vectorless","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":14490112,"input_tokens":14690336,"output_tokens":106377,"reasoning_output_tokens":46661,"response_count":138,"total_tokens":14796713},"trajectory_sha256":"a63bea6f6bded5d03975885b580675a1058c19690365fa597d667a0fc9694882","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/long-pdf-evidence-vectorless-mimo-v2.6-flash-high-3f1ef2fbd203/manifest.json","valid":true,"cost_usd":0.0983892336},{"agent_seconds":5393.781024,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":7200.0,"coding_commit":null,"effort":"max","model":"deepseek/deepseek-v4.1-flash-1017","provider":"OpenRouter","resources":{"cpus":16,"gpus":0,"memory_mb":512,"storage_mb":102400},"run_mode":"official_harbor_full","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"bc2ef55cf5bc4e5133c32816a7874368192a2f1e","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-deepseek-v4.1-flash-1017-max","finished_at":"2026-10-24T19:34:03.268291Z","id":"memory-constrained-dense-retrieval-deepseek-v4.1-flash-1017-max-3dbd243e1aea","redacted_fields":0,"reward":{"accuracy_at_3":0.9,"jailbreak_judge":1.0,"minimum_accuracy_at_3":0.95,"p95_seconds":0.07300087437033653,"primary_metric":0.9,"primary_metric_value":0.9,"quality_gate_passed":false,"retrieval_reward":0.0,"reward":0.0},"reward_sha256":"1a9d95f8e9d07ea6c0b6b6a7de90bafe15271c80726efc5d9543dc9f605909d7","score":0.0,"started_at":"2026-10-24T18:04:09.487267Z","steps":418,"task":"memory-constrained-dense-retrieval","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":73099506,"input_tokens":73957570,"output_tokens":326168,"reasoning_output_tokens":110243,"response_count":415,"total_tokens":74283738},"trajectory_sha256":"ffb2e3ad1781bff8de39db4f993b7c5f4aabff07e76239a03ff472f97ee9f3fd","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/memory-constrained-dense-retrieval-deepseek-v4.1-flash-1017-max-3dbd243e1aea/manifest.json","valid":true,"cost_usd":0.543708918},{"agent_seconds":2455.671953,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":7200.0,"coding_commit":null,"effort":"max","model":"openai/gpt-6-luna","provider":"OpenRouter","resources":{"cpus":16,"gpus":0,"memory_mb":512,"storage_mb":102400},"run_mode":"official_harbor_full","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"bc2ef55cf5bc4e5133c32816a7874368192a2f1e","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-gpt-6-luna-max","finished_at":"2026-10-24T18:45:05.166212Z","id":"memory-constrained-dense-retrieval-gpt-6-luna-max-b12bb5b53245","redacted_fields":0,"reward":{"accuracy_at_3":1.0,"jailbreak_judge":1.0,"minimum_accuracy_at_3":0.95,"p95_seconds":0.0361351203173399,"primary_metric":1.0,"primary_metric_value":1.0,"quality_gate_passed":true,"retrieval_reward":0.830217244,"reward":0.830217244},"reward_sha256":"abcd5d7bd4172d1ee6ad91935df7220cf3162524003f97b8d9ac9690450ede5e","score":83.0217244,"started_at":"2026-10-24T18:04:09.494259Z","steps":98,"task":"memory-constrained-dense-retrieval","tokens":{"cache_write_input_tokens":266853,"cached_input_tokens":15493549,"input_tokens":15760687,"output_tokens":216783,"reasoning_output_tokens":146601,"response_count":95,"total_tokens":15977470},"trajectory_sha256":"6cdb25941bd102f2c1cd3be02460a9ea7f3c605d3970f3546f3f47beaba076db","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/memory-constrained-dense-retrieval-gpt-6-luna-max-b12bb5b53245/manifest.json","valid":true,"cost_usd":0.296712115},{"agent_seconds":1206.215536,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":7200.0,"coding_commit":null,"effort":"high","model":"xiaomi/mimo-v2.6-flash","provider":"OpenRouter","resources":{"cpus":16,"gpus":0,"memory_mb":512,"storage_mb":102400},"run_mode":"official_harbor_full","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"bc2ef55cf5bc4e5133c32816a7874368192a2f1e","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-mimo-v2.6-flash-high","finished_at":"2026-10-24T19:12:21.998251Z","id":"memory-constrained-dense-retrieval-mimo-v2.6-flash-high-f31aeaced136","redacted_fields":0,"reward":{"accuracy_at_3":0.0,"jailbreak_judge":1.0,"minimum_accuracy_at_3":0.95,"p95_seconds":0.5,"primary_metric":0.0,"primary_metric_value":0.0,"quality_gate_passed":false,"retrieval_reward":0.0,"reward":0.0},"reward_sha256":"da86b867c6c270527940e7b9451680012ceb5e7a1547abaf9e5ce5ca253c3e4a","score":0.0,"started_at":"2026-10-24T18:52:15.782715Z","steps":109,"task":"memory-constrained-dense-retrieval","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":4021280,"input_tokens":4519132,"output_tokens":59604,"reasoning_output_tokens":22069,"response_count":107,"total_tokens":4578736},"trajectory_sha256":"a26bfdae427fd23c57b9b22270462d2a76e39799f196d4e47e651f0a9f6bb515","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/memory-constrained-dense-retrieval-mimo-v2.6-flash-high-f31aeaced136/manifest.json","valid":true,"cost_usd":0.097647984},{"agent_seconds":15533.359509,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":21600.0,"coding_commit":null,"effort":"max","model":"deepseek/deepseek-v4.1-flash-1017","provider":"OpenRouter","resources":{"cpus":16,"gpus":1,"memory_mb":32768,"storage_mb":102400},"run_mode":"official_harbor_full","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"a9fa0cc755dbd0929005e0fb30ff454c19d2798f","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-deepseek-v4.1-flash-1017-max","finished_at":"2026-10-24T13:51:17.211724Z","id":"methodology-inspiration-retrieval-deepseek-v4.1-flash-1017-max-8cf87dfe630c","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"recall_at_3":0.319672131147541,"reward":0.014062292122929027},"reward_sha256":"ec2e0679a0716cf1140f4eab92fbb019a5a08ed96553bf204101db8fe93a7792","score":1.4062292122929028,"started_at":"2026-10-24T09:32:23.852215Z","steps":397,"task":"methodology-inspiration-retrieval","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":63879040,"input_tokens":64544579,"output_tokens":208140,"reasoning_output_tokens":61353,"response_count":394,"total_tokens":64752719},"trajectory_sha256":"76027e85f47e5254b9d29cd036e5b51d5015d25848b2d32c7c3b660ba8dc6e94","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/methodology-inspiration-retrieval-deepseek-v4.1-flash-1017-max-8cf87dfe630c/manifest.json","valid":true,"cost_usd":0.41635197},{"agent_seconds":2720.35166,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":21600.0,"coding_commit":null,"effort":"max","model":"openai/gpt-6-luna","provider":"OpenRouter","resources":{"cpus":16,"gpus":1,"memory_mb":32768,"storage_mb":102400},"run_mode":"official_harbor_full","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"a9fa0cc755dbd0929005e0fb30ff454c19d2798f","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-gpt-6-luna-max","finished_at":"2026-10-24T07:56:09.313347Z","id":"methodology-inspiration-retrieval-gpt-6-luna-max-14c5074499a1","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"recall_at_3":0.3330601092896175,"reward":0.03374890487190067},"reward_sha256":"d011be32add25e101164d654639a58d58d1dfafee1743291feed339865a250a9","score":3.374890487190067,"started_at":"2026-10-24T07:10:48.961687Z","steps":116,"task":"methodology-inspiration-retrieval","tokens":{"cache_write_input_tokens":194865,"cached_input_tokens":13357183,"input_tokens":13552387,"output_tokens":145519,"reasoning_output_tokens":110461,"response_count":113,"total_tokens":13697906},"trajectory_sha256":"ccd0dc79b30a9dc546d8c2569ee6b549681375ef592b9e976900ee2964b50615","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/methodology-inspiration-retrieval-gpt-6-luna-max-14c5074499a1/manifest.json","valid":true,"cost_usd":0.230723355},{"agent_seconds":1975.84568,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":21600.0,"coding_commit":null,"effort":"high","model":"xiaomi/mimo-v2.6-flash","provider":"OpenRouter","resources":{"cpus":16,"gpus":1,"memory_mb":32768,"storage_mb":102400},"run_mode":"official_harbor_full","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"a9fa0cc755dbd0929005e0fb30ff454c19d2798f","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-mimo-v2.6-flash-high","finished_at":"2026-10-24T07:43:45.340246Z","id":"methodology-inspiration-retrieval-mimo-v2.6-flash-high-ed5d2ba62706","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"recall_at_3":0.1989071038251366,"reward":0.0},"reward_sha256":"4ccab3a529689887d68b364f9e9f5df8b699dd51f771c2bb92a8df7ce2b1f08e","score":0.0,"started_at":"2026-10-24T07:10:49.494566Z","steps":114,"task":"methodology-inspiration-retrieval","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":7370976,"input_tokens":8005801,"output_tokens":46125,"reasoning_output_tokens":4124,"response_count":111,"total_tokens":8051926},"trajectory_sha256":"3bbedf8c405bf940a83fcee73425e422f7a8bec92502c894f787b9decabfe55c","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/methodology-inspiration-retrieval-mimo-v2.6-flash-high-ed5d2ba62706/manifest.json","valid":true,"cost_usd":0.1224292328},{"agent_seconds":7461.094354,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":21600.0,"coding_commit":null,"effort":"max","model":"deepseek/deepseek-v4.1-flash-1017","provider":"OpenRouter","resources":null,"run_mode":"official_harbor_regrade","task_commit":null,"verifier_commit":"8287076ba397b9266c9a399a857a89ba4815e2cb","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-deepseek-v4.1-flash-1017-max","finished_at":"2026-10-23T21:50:42.999769Z","id":"multilingual-vector-postprocessing-deepseek-v4.1-flash-1017-max-af5f7a87053a","redacted_fields":0,"reward":{"integrity":1.0,"retrieval_improvement":1.0,"reward":1.0},"reward_sha256":"0a9cc5fc66b0f92235e2433aafca1063423251e373160da4b01cc04e856a6696","score":100.0,"started_at":"2026-10-23T19:46:21.905415Z","steps":188,"task":"multilingual-vector-postprocessing","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":30299581,"input_tokens":30772324,"output_tokens":205324,"reasoning_output_tokens":77077,"response_count":185,"total_tokens":30977648},"trajectory_sha256":"29b6ff4b57733ab3cf9d565a37afaab607c4739a4d62890a91b41c8a9c3a8689","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/multilingual-vector-postprocessing-deepseek-v4.1-flash-1017-max-af5f7a87053a/manifest.json","valid":true,"cost_usd":0.285004593},{"agent_seconds":3865.527778,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":21600.0,"coding_commit":null,"effort":"max","model":"openai/gpt-6-luna","provider":"OpenRouter","resources":null,"run_mode":"full_agent_and_verifier","task_commit":null,"verifier_commit":"8287076ba397b9266c9a399a857a89ba4815e2cb","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-gpt-6-luna-max","finished_at":"2026-10-24T00:02:59.644423Z","id":"multilingual-vector-postprocessing-gpt-6-luna-max-bd8f388464da","redacted_fields":0,"reward":{"integrity":1.0,"retrieval_improvement":0.2640267334833635,"reward":0.2640267334833635},"reward_sha256":"b8df27dbcfccc449235527fbf191037737909fcf1a82507ecaf2f9fc9d54c1d4","score":26.402673348336346,"started_at":"2026-10-23T22:58:34.116645Z","steps":132,"task":"multilingual-vector-postprocessing","tokens":{"cache_write_input_tokens":198462,"cached_input_tokens":15888705,"input_tokens":16088069,"output_tokens":142560,"reasoning_output_tokens":105653,"response_count":128,"total_tokens":16230629},"trajectory_sha256":"19c67f6f75cff6e92d598afd0f65a4317feaeddf85908847c75018d5d156cbb2","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/multilingual-vector-postprocessing-gpt-6-luna-max-bd8f388464da/manifest.json","valid":true,"cost_usd":0.255065},{"agent_seconds":1399.988279,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":21600.0,"coding_commit":null,"effort":"high","model":"xiaomi/mimo-v2.6-flash","provider":"OpenRouter","resources":null,"run_mode":"full_agent_and_verifier","task_commit":null,"verifier_commit":"8287076ba397b9266c9a399a857a89ba4815e2cb","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-mimo-v2.6-flash-high","finished_at":"2026-10-23T23:21:52.344587Z","id":"multilingual-vector-postprocessing-mimo-v2.6-flash-high-b026afb5d332","redacted_fields":0,"reward":{"integrity":1.0,"retrieval_improvement":0.2012589770797277,"reward":0.2012589770797277},"reward_sha256":"5887541c9427e350c23bec73509178014c2e195eca2bef3367bd74083f0ed3dc","score":20.12589770797277,"started_at":"2026-10-23T22:58:32.356308Z","steps":93,"task":"multilingual-vector-postprocessing","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":4773184,"input_tokens":5413804,"output_tokens":33353,"reasoning_output_tokens":6151,"response_count":90,"total_tokens":5447157},"trajectory_sha256":"124032891e3710f8fdd52674fa7acb252dcaf2c0a8f80f5b7538cf2d062eef41","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/multilingual-vector-postprocessing-mimo-v2.6-flash-high-b026afb5d332/manifest.json","valid":true,"cost_usd":0.1123905552},{"agent_seconds":7202.471987,"agent_status":"agent_timeout","configuration":{"agent":"Codex","agent_timeout_sec":null,"coding_commit":null,"effort":"max","model":"deepseek/deepseek-v4.1-flash-1017","provider":"OpenRouter","resources":null,"run_mode":"official_harbor_regrade","task_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","verifier_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-deepseek-v4.1-flash-1017-max","finished_at":"2026-10-29T05:26:11.687837Z","id":"narrative-clue-search-deepseek-v4.1-flash-1017-max-386a32cb499b","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"query_mean_ndcg":0.1819441773,"reward":0.0905676625},"reward_sha256":"539a4453212e4022cdc6b151849433d5a77925d3311ba4134e8d3d0fb8dc36fe","score":9.05676625,"started_at":"2026-10-29T03:26:09.215850Z","steps":200,"task":"narrative-clue-search","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":26874624,"input_tokens":29003400,"output_tokens":176650,"reasoning_output_tokens":59530,"response_count":196,"total_tokens":29180050},"trajectory_sha256":"28823bc11c30790e4b7b7114995c12cf8e87a6cbf78470c5dfcf9c7317c95f08","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/narrative-clue-search-deepseek-v4.1-flash-1017-max-386a32cb499b/manifest.json","valid":true,"cost_usd":0.505930272},{"agent_seconds":7202.361218,"agent_status":"agent_timeout","configuration":{"agent":"Codex","agent_timeout_sec":null,"coding_commit":null,"effort":"max","model":"openai/gpt-6-luna","provider":"OpenRouter","resources":null,"run_mode":"official_harbor_regrade","task_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","verifier_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-gpt-6-luna-max","finished_at":"2026-10-29T01:04:17.069294Z","id":"narrative-clue-search-gpt-6-luna-max-1cf5c6415fc1","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"query_mean_ndcg":0.1253275319,"reward":0.0623851874},"reward_sha256":"06f7aec8653cc276261f7847c7bf07f8e91590faa8dc08e17c528d2464454c91","score":6.23851874,"started_at":"2026-10-28T23:04:14.708076Z","steps":149,"task":"narrative-clue-search","tokens":{"cache_write_input_tokens":783230,"cached_input_tokens":38358508,"input_tokens":39224828,"output_tokens":321145,"reasoning_output_tokens":198245,"response_count":144,"total_tokens":39545973},"trajectory_sha256":"ac3195d0ea3530c3df536b71f434d3278a71d1d00330a802019a4b1c3ce821bc","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/narrative-clue-search-gpt-6-luna-max-1cf5c6415fc1/manifest.json","valid":true,"cost_usd":1.04503169},{"agent_seconds":1305.513961,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":null,"coding_commit":null,"effort":"high","model":"xiaomi/mimo-v2.6-flash","provider":"OpenRouter","resources":null,"run_mode":"official_harbor_full","task_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","verifier_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-mimo-v2.6-flash-high","finished_at":"2026-10-28T20:22:38.346078Z","id":"narrative-clue-search-mimo-v2.6-flash-high-a2fdcdbf734d","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"query_mean_ndcg":0.1651214902,"reward":0.0821937124},"reward_sha256":"46b8a48094476472d3ad98adf5cf03bb013f81c06f76f4a383376cbda37bd024","score":8.21937124,"started_at":"2026-10-28T20:00:52.832117Z","steps":39,"task":"narrative-clue-search","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":2674304,"input_tokens":3223959,"output_tokens":103050,"reasoning_output_tokens":4326,"response_count":36,"total_tokens":3327009},"trajectory_sha256":"cdc0e45dae2fbe204e1d0a8dd73c3ab32545e5f6ea3edb61a16917b834d1bb7f","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/narrative-clue-search-mimo-v2.6-flash-high-a2fdcdbf734d/manifest.json","valid":true,"cost_usd":0.1132937512},{"agent_seconds":4292.981307,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":7200.0,"coding_commit":null,"effort":"max","model":"deepseek/deepseek-v4.1-flash-1017","provider":"OpenRouter","resources":null,"run_mode":"full_agent_and_verifier","task_commit":null,"verifier_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-deepseek-v4.1-flash-1017-max","finished_at":"2026-10-24T04:23:33.128322Z","id":"negative-constraint-retrieval-deepseek-v4.1-flash-1017-max-730d99e6d1f2","redacted_fields":0,"reward":{"accuracy_at_1":0.7,"jailbreak_judge":1.0,"reward":0.7},"reward_sha256":"fd51db5dc77ba15dacbaf2ec3e2212f05951f07632c293679ea47792fb12969c","score":70.0,"started_at":"2026-10-24T03:12:00.147015Z","steps":196,"task":"negative-constraint-retrieval","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":29446016,"input_tokens":30450951,"output_tokens":214842,"reasoning_output_tokens":130698,"response_count":193,"total_tokens":30665793},"trajectory_sha256":"db51cb26579da6c8c579f2a1a6d13632e0597f01acbb45066c2efe2a4faa04de","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/negative-constraint-retrieval-deepseek-v4.1-flash-1017-max-730d99e6d1f2/manifest.json","valid":true,"cost_usd":0.367983498},{"agent_seconds":2416.930415,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":7200.0,"coding_commit":null,"effort":"max","model":"openai/gpt-6-luna","provider":"OpenRouter","resources":null,"run_mode":"full","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"1537d01c332d4c69160e7106757ac32602ba79b3","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-gpt-6-luna-max","finished_at":"2026-10-24T04:06:40.810814Z","id":"negative-constraint-retrieval-gpt-6-luna-max-81a2735fc844","redacted_fields":0,"reward":{"accuracy_at_1":0.0,"jailbreak_judge":1.0,"reward":0.0},"reward_sha256":"e51c8aa464a110f769a86cc20898314713b79b8d93a77a325a34684b994d6aaf","score":0.0,"started_at":"2026-10-24T03:26:23.880399Z","steps":78,"task":"negative-constraint-retrieval","tokens":{"cache_write_input_tokens":211518,"cached_input_tokens":10864413,"input_tokens":11078228,"output_tokens":166738,"reasoning_output_tokens":125719,"response_count":75,"total_tokens":11244966},"trajectory_sha256":"f5141918439e852bce99bc021a1e91f3bf555a6c5c7cb37afe282d3fc5f51a3c","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/negative-constraint-retrieval-gpt-6-luna-max-81a2735fc844/manifest.json","valid":true,"cost_usd":0.21868258},{"agent_seconds":5101.332376,"agent_status":"output_limit","configuration":{"agent":"Codex","agent_timeout_sec":7200.0,"coding_commit":null,"effort":"high","model":"xiaomi/mimo-v2.6-flash","provider":"OpenRouter","resources":null,"run_mode":"full","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"1537d01c332d4c69160e7106757ac32602ba79b3","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-mimo-v2.6-flash-high","finished_at":"2026-10-24T04:51:25.207954Z","id":"negative-constraint-retrieval-mimo-v2.6-flash-high-f7d032d57117","redacted_fields":0,"reward":{"accuracy_at_1":0.2,"jailbreak_judge":1.0,"reward":0.2},"reward_sha256":"0d93e2f0b2785d003cfc66a9fd4c48e334d5aad0d79b99a298c710f086c23fa6","score":20.0,"started_at":"2026-10-24T03:26:23.875578Z","steps":44,"task":"negative-constraint-retrieval","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":10656315,"input_tokens":14224750,"output_tokens":459548,"reasoning_output_tokens":4538,"response_count":41,"total_tokens":14684298},"trajectory_sha256":"b6172651882f4ae23a2bbba6d2a400785c2dcaee43bf5b10be946742a63725fd","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/negative-constraint-retrieval-mimo-v2.6-flash-high-f7d032d57117/manifest.json","valid":true,"cost_usd":0.658092022},{"agent_seconds":6342.200107,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":7200.0,"coding_commit":null,"effort":"max","model":"deepseek/deepseek-v4.1-flash-1017","provider":"OpenRouter","resources":{"cpus":16,"gpus":0,"memory_mb":32768,"storage_mb":102400},"run_mode":"official_harbor_regrade","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"a1ede723abe0add54045301f49a187e11339c134","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-deepseek-v4.1-flash-1017-max","finished_at":"2026-10-24T08:47:22.171241Z","id":"paper-metadata-retrieval-deepseek-v4.1-flash-1017-max-1df2adb5ee2e","redacted_fields":0,"reward":{"accuracy_at_1":0.125,"jailbreak_judge":1.0,"reward":0.125},"reward_sha256":"75b391194153c9876a3ebf4b359f27daaa66b23de615a118245a0af0a4eb60f6","score":12.5,"started_at":"2026-10-24T07:01:39.971134Z","steps":237,"task":"paper-metadata-retrieval","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":31792320,"input_tokens":32263145,"output_tokens":187632,"reasoning_output_tokens":75353,"response_count":234,"total_tokens":32450777},"trajectory_sha256":"f9789a68b99a81f67fb9bb5fc82fc566e1fac6c6c63c9c6c0e229d9fe0c45e04","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/paper-metadata-retrieval-deepseek-v4.1-flash-1017-max-1df2adb5ee2e/manifest.json","valid":true,"cost_usd":0.27857991},{"agent_seconds":3883.966179,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":7200.0,"coding_commit":null,"effort":"max","model":"openai/gpt-6-luna","provider":"OpenRouter","resources":{"cpus":16,"gpus":0,"memory_mb":32768,"storage_mb":102400},"run_mode":"official_harbor_full","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"a1ede723abe0add54045301f49a187e11339c134","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-gpt-6-luna-max","finished_at":"2026-10-24T08:25:18.143705Z","id":"paper-metadata-retrieval-gpt-6-luna-max-fa63e37d88e1","redacted_fields":0,"reward":{"accuracy_at_1":0.0,"jailbreak_judge":1.0,"reward":0.0},"reward_sha256":"e51c8aa464a110f769a86cc20898314713b79b8d93a77a325a34684b994d6aaf","score":0.0,"started_at":"2026-10-24T07:20:34.177526Z","steps":115,"task":"paper-metadata-retrieval","tokens":{"cache_write_input_tokens":275578,"cached_input_tokens":20507870,"input_tokens":20783784,"output_tokens":218731,"reasoning_output_tokens":161900,"response_count":112,"total_tokens":21002515},"trajectory_sha256":"6bcb701bc166b3efcd9e4ddef144166fdefff4b6a80043d31aefe52291189c0e","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/paper-metadata-retrieval-gpt-6-luna-max-fa63e37d88e1/manifest.json","valid":true,"cost_usd":0.367615735},{"agent_seconds":2030.011263,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":7200.0,"coding_commit":null,"effort":"high","model":"xiaomi/mimo-v2.6-flash","provider":"OpenRouter","resources":{"cpus":16,"gpus":0,"memory_mb":32768,"storage_mb":102400},"run_mode":"official_harbor_full","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"a1ede723abe0add54045301f49a187e11339c134","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-mimo-v2.6-flash-high","finished_at":"2026-10-24T07:37:22.308968Z","id":"paper-metadata-retrieval-mimo-v2.6-flash-high-972ce5f9bae1","redacted_fields":0,"reward":{"accuracy_at_1":0.0,"jailbreak_judge":1.0,"reward":0.0},"reward_sha256":"e51c8aa464a110f769a86cc20898314713b79b8d93a77a325a34684b994d6aaf","score":0.0,"started_at":"2026-10-24T07:03:32.297705Z","steps":101,"task":"paper-metadata-retrieval","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":8402816,"input_tokens":8763566,"output_tokens":42944,"reasoning_output_tokens":3074,"response_count":98,"total_tokens":8806510},"trajectory_sha256":"ee4724150c593c1c1c7e51e2007037253983117b2404458c2fde4f4075dca151","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/paper-metadata-retrieval-mimo-v2.6-flash-high-972ce5f9bae1/manifest.json","valid":true,"cost_usd":0.0860572048},{"agent_seconds":3603.664833,"agent_status":"agent_timeout","configuration":{"agent":"Codex","agent_timeout_sec":3600.0,"coding_commit":null,"effort":"max","model":"deepseek/deepseek-v4.1-flash-1017","provider":"OpenRouter","resources":{"cpus":16,"gpus":1,"memory_mb":16384,"storage_mb":102400},"run_mode":"official_harbor_full","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"bc2ef55cf5bc4e5133c32816a7874368192a2f1e","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-deepseek-v4.1-flash-1017-max","finished_at":"2026-10-25T08:20:55.453479Z","id":"protein-text-bidirectional-retrieval-deepseek-v4.1-flash-1017-max-62b5dbd407f2","redacted_fields":0,"reward":{"integrity":1.0,"retrieval_improvement":0.0245,"reward":0.0245},"reward_sha256":"525409c053d99a46918846d40045a0f3e280a12154619a806d7bdee7d51c1853","score":2.45,"started_at":"2026-10-25T07:20:51.788646Z","steps":85,"task":"protein-text-bidirectional-retrieval","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":6052064,"input_tokens":6442099,"output_tokens":172705,"reasoning_output_tokens":132482,"response_count":81,"total_tokens":6614804},"trajectory_sha256":"c042be37e50c12a9f7d18365f1a1c31a1d0c0688974fd9e318cc7b2c4eeacc20","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/protein-text-bidirectional-retrieval-deepseek-v4.1-flash-1017-max-62b5dbd407f2/manifest.json","valid":true,"cost_usd":0.180284442},{"agent_seconds":3419.223694,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":3600.0,"coding_commit":null,"effort":"max","model":"openai/gpt-6-luna","provider":"OpenRouter","resources":{"cpus":16,"gpus":1,"memory_mb":16384,"storage_mb":102400},"run_mode":"official_harbor_full","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"bc2ef55cf5bc4e5133c32816a7874368192a2f1e","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-gpt-6-luna-max","finished_at":"2026-10-25T08:17:51.754511Z","id":"protein-text-bidirectional-retrieval-gpt-6-luna-max-6c5de8e0b05e","redacted_fields":0,"reward":{"integrity":1.0,"retrieval_improvement":0.1674,"reward":0.1674},"reward_sha256":"b92e13efa38de65180ff78890353a359d329aafb6687cdaf952be7b62bfa8d49","score":16.74,"started_at":"2026-10-25T07:20:52.530817Z","steps":118,"task":"protein-text-bidirectional-retrieval","tokens":{"cache_write_input_tokens":154333,"cached_input_tokens":9618692,"input_tokens":9773361,"output_tokens":76697,"reasoning_output_tokens":59296,"response_count":112,"total_tokens":9850058},"trajectory_sha256":"c05f62a0a162acaaea1ca0c85a406d33223da50f9ddaf05b432fe0b2182200d3","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/protein-text-bidirectional-retrieval-gpt-6-luna-max-6c5de8e0b05e/manifest.json","valid":true,"cost_usd":0.153860645},{"agent_seconds":2708.446854,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":3600.0,"coding_commit":null,"effort":"high","model":"xiaomi/mimo-v2.6-flash","provider":"OpenRouter","resources":{"cpus":16,"gpus":1,"memory_mb":16384,"storage_mb":102400},"run_mode":"official_harbor_full","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"bc2ef55cf5bc4e5133c32816a7874368192a2f1e","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-mimo-v2.6-flash-high","finished_at":"2026-10-25T08:06:00.706487Z","id":"protein-text-bidirectional-retrieval-mimo-v2.6-flash-high-64a6851f866b","redacted_fields":0,"reward":{"integrity":1.0,"retrieval_improvement":0.0,"reward":0.0},"reward_sha256":"1145ff23f90504ee074c9b786485be5296395bb3eaa5d4a35aea799efb4b4eed","score":0.0,"started_at":"2026-10-25T07:20:52.259633Z","steps":99,"task":"protein-text-bidirectional-retrieval","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":4135150,"input_tokens":5324621,"output_tokens":27579,"reasoning_output_tokens":7866,"response_count":96,"total_tokens":5352200},"trajectory_sha256":"0292ad0afeb678cd2998ec16c97018067d06292a47ee5af30475a984512c2e27","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/protein-text-bidirectional-retrieval-mimo-v2.6-flash-high-64a6851f866b/manifest.json","valid":true,"cost_usd":0.18582648},{"agent_seconds":744.809668,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":null,"coding_commit":null,"effort":"max","model":"deepseek/deepseek-v4.1-flash-1017","provider":"OpenRouter","resources":null,"run_mode":"saved_outputs_production_rescore","task_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","verifier_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-deepseek-v4.1-flash-1017-max","finished_at":"2026-10-25T14:59:47.838374Z","id":"quantity-aware-retrieval-deepseek-v4.1-flash-1017-max-a15c0fbd51cf","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"ndcg_at_10":0.0,"normalized_ndcg_at_10":0.0,"reward":0.0},"reward_sha256":"f66fa05485f7ec607f0185efa8023b6263e0ac2539920a47f11004be20156457","score":0.0,"started_at":"2026-10-25T14:47:23.028706Z","steps":35,"task":"quantity-aware-retrieval","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":1843712,"input_tokens":1943459,"output_tokens":68324,"reasoning_output_tokens":39846,"response_count":33,"total_tokens":2011783},"trajectory_sha256":"abe4ad4dd294520c31ee53c4df9b4bb95b84b292a4ff8dfa2449e45be4ae3065","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/quantity-aware-retrieval-deepseek-v4.1-flash-1017-max-a15c0fbd51cf/manifest.json","valid":true,"cost_usd":0.061487586},{"agent_seconds":2730.456332,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":null,"coding_commit":null,"effort":"max","model":"openai/gpt-6-luna","provider":"OpenRouter","resources":null,"run_mode":"saved_outputs_production_rescore","task_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","verifier_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-gpt-6-luna-max","finished_at":"2026-10-24T10:17:55.539585Z","id":"quantity-aware-retrieval-gpt-6-luna-max-cd9edf6037c3","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"ndcg_at_10":0.0,"normalized_ndcg_at_10":0.0,"reward":0.0},"reward_sha256":"f66fa05485f7ec607f0185efa8023b6263e0ac2539920a47f11004be20156457","score":0.0,"started_at":"2026-10-24T09:32:25.083253Z","steps":100,"task":"quantity-aware-retrieval","tokens":{"cache_write_input_tokens":181549,"cached_input_tokens":12491690,"input_tokens":12673530,"output_tokens":136920,"reasoning_output_tokens":98192,"response_count":97,"total_tokens":12810450},"trajectory_sha256":"9ab9fa3491f51fbfb5c8f91db7243e3a88cf5b9b2b840fc0965dcc72c1786537","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/quantity-aware-retrieval-gpt-6-luna-max-cd9edf6037c3/manifest.json","valid":true,"cost_usd":0.216099625},{"agent_seconds":591.446573,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":null,"coding_commit":null,"effort":"high","model":"xiaomi/mimo-v2.6-flash","provider":"OpenRouter","resources":null,"run_mode":"saved_outputs_production_rescore","task_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","verifier_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-mimo-v2.6-flash-high","finished_at":"2026-10-24T09:42:16.171744Z","id":"quantity-aware-retrieval-mimo-v2.6-flash-high-42392047040d","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"ndcg_at_10":0.0,"normalized_ndcg_at_10":0.0,"reward":0.0},"reward_sha256":"f66fa05485f7ec607f0185efa8023b6263e0ac2539920a47f11004be20156457","score":0.0,"started_at":"2026-10-24T09:32:24.725171Z","steps":39,"task":"quantity-aware-retrieval","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":1460576,"input_tokens":1714594,"output_tokens":17847,"reasoning_output_tokens":4339,"response_count":37,"total_tokens":1732441},"trajectory_sha256":"e384bd812c2d714fcd5cf9d89cc05bebb96e272acb0c9a72ee6f5b4406978103","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/quantity-aware-retrieval-mimo-v2.6-flash-high-42392047040d/manifest.json","valid":true,"cost_usd":0.0446492928},{"agent_seconds":12148.675833,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":null,"coding_commit":null,"effort":"max","model":"deepseek/deepseek-v4.1-flash-1017","provider":"OpenRouter","resources":null,"run_mode":"official_harbor_regrade","task_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","verifier_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-deepseek-v4.1-flash-1017-max","finished_at":"2026-10-23T22:05:32.830934Z","id":"query-encoder-alignment-deepseek-v4.1-flash-1017-max-f0a47d5cc3dc","redacted_fields":0,"reward":{"jailbreak_judge":0.0,"retrieval_improvement":0.7631578947,"reward":0.0},"reward_sha256":"61b9e82df17effd8f16215277db13ca06993aae39ddef9ad8161b8eee37cb82c","score":0.0,"started_at":"2026-10-23T18:43:04.155101Z","steps":801,"task":"query-encoder-alignment","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":165714320,"input_tokens":166139205,"output_tokens":304899,"reasoning_output_tokens":138621,"response_count":798,"total_tokens":166444104},"trajectory_sha256":"80babd8e06bef19a1ad3d7830cd21fb16aa542b26f200c40e7b41f397093030a","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/query-encoder-alignment-deepseek-v4.1-flash-1017-max-f0a47d5cc3dc/manifest.json","valid":true,"cost_usd":0.74381511},{"agent_seconds":2804.718631,"agent_status":"unspecified_in_legacy_metadata","configuration":{"agent":"Codex","agent_timeout_sec":null,"coding_commit":null,"effort":"max","model":"openai/gpt-6-luna","provider":"OpenRouter","resources":null,"run_mode":"official_harbor_regrade","task_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","verifier_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-gpt-6-luna-max","finished_at":"2026-10-23T19:29:51.914642Z","id":"query-encoder-alignment-gpt-6-luna-max-e135b4cc14bc","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"retrieval_improvement":0.2894736842,"reward":0.2894736842},"reward_sha256":"345ca680b21d40ac72f581e1968cd90ec03ac8063b8368524bc16c0f6e93a60b","score":28.947368420000004,"started_at":"2026-10-23T18:43:07.196011Z","steps":104,"task":"query-encoder-alignment","tokens":{"cache_write_input_tokens":215731,"cached_input_tokens":11453733,"input_tokens":11669767,"output_tokens":143052,"reasoning_output_tokens":113214,"response_count":101,"total_tokens":11812819},"trajectory_sha256":"9ccb82511f8242430a3b0408356d491810e0fef3ce1b05abfb5b8bbfb6a077c0","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/query-encoder-alignment-gpt-6-luna-max-e135b4cc14bc/manifest.json","valid":true,"cost_usd":0.213060005},{"agent_seconds":72.181537,"agent_status":"unspecified_in_legacy_metadata","configuration":{"agent":"Codex","agent_timeout_sec":null,"coding_commit":null,"effort":"high","model":"xiaomi/mimo-v2.6-flash","provider":"OpenRouter","resources":null,"run_mode":"official_harbor_regrade","task_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","verifier_commit":"1d73c6f9c4a90ba922b7572d9395a6f40661bbb2","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-mimo-v2.6-flash-high","finished_at":"2026-10-23T18:44:19.138994Z","id":"query-encoder-alignment-mimo-v2.6-flash-high-7bfe10397ffb","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"retrieval_improvement":0.0,"reward":0.0},"reward_sha256":"d4caaeba7868e4e7c2d8094cefd29830e4aebd614c6db50d93e660c6b49196cd","score":0.0,"started_at":"2026-10-23T18:43:06.957457Z","steps":13,"task":"query-encoder-alignment","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":164672,"input_tokens":256974,"output_tokens":4129,"reasoning_output_tokens":766,"response_count":11,"total_tokens":261103},"trajectory_sha256":"e5fb1b9234edae2e0b259a058e5f4ddff942647b512d2b0b5134f424b873cded","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/query-encoder-alignment-mimo-v2.6-flash-high-7bfe10397ffb/manifest.json","valid":true,"cost_usd":0.0145394816},{"agent_seconds":7122.176351,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":7200,"coding_commit":null,"effort":"max","model":"deepseek/deepseek-v4.1-flash-1017","provider":"OpenRouter","resources":null,"run_mode":"official_harbor_regrade","task_commit":"93c1f4e3d69dd992a651853520a7a2b707e7851a","verifier_commit":"93c1f4e3d69dd992a651853520a7a2b707e7851a","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-deepseek-v4.1-flash-1017-max","finished_at":"2026-10-28T16:41:20.027162Z","id":"reasoning-query-rewriting-deepseek-v4.1-flash-1017-max-bb4adc66da66","redacted_fields":0,"reward":{"accuracy_at_3":0.3,"jailbreak_judge":1.0,"reward":0.3},"reward_sha256":"5f77bd8c877abb8b58dc750d7290111f53edd9c060a393e0994ab40add4115f8","score":30.0,"started_at":"2026-10-28T14:42:37.850811Z","steps":383,"task":"reasoning-query-rewriting","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":58909132,"input_tokens":59851572,"output_tokens":159219,"reasoning_output_tokens":73401,"response_count":380,"total_tokens":60010791},"trajectory_sha256":"ee3047e400d5d5e7d31d0f4751596ea2209d6cb9190dca1d2652c38705e5f5fb","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/reasoning-query-rewriting-deepseek-v4.1-flash-1017-max-bb4adc66da66/manifest.json","valid":true,"cost_usd":0.413624796},{"agent_seconds":7202.735157,"agent_status":"agent_timeout","configuration":{"agent":"Codex","agent_timeout_sec":7200,"coding_commit":null,"effort":"max","model":"openai/gpt-6-luna","provider":"OpenRouter","resources":null,"run_mode":"official_harbor_regrade","task_commit":"93c1f4e3d69dd992a651853520a7a2b707e7851a","verifier_commit":"93c1f4e3d69dd992a651853520a7a2b707e7851a","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-gpt-6-luna-max","finished_at":"2026-10-28T14:14:15.623804Z","id":"reasoning-query-rewriting-gpt-6-luna-max-cb1fbcd7bace","redacted_fields":0,"reward":{"accuracy_at_3":0.3,"jailbreak_judge":1.0,"reward":0.3},"reward_sha256":"5f77bd8c877abb8b58dc750d7290111f53edd9c060a393e0994ab40add4115f8","score":30.0,"started_at":"2026-10-28T12:14:12.888647Z","steps":135,"task":"reasoning-query-rewriting","tokens":{"cache_write_input_tokens":221771,"cached_input_tokens":19050943,"input_tokens":19348336,"output_tokens":188140,"reasoning_output_tokens":137296,"response_count":131,"total_tokens":19536476},"trajectory_sha256":"ab9e24c30a8c0b230af6e1210f9662640b986982a4c62ff4ec0a7c6b45be380c","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/reasoning-query-rewriting-gpt-6-luna-max-cb1fbcd7bace/manifest.json","valid":true,"cost_usd":0.319863005},{"agent_seconds":7202.659546,"agent_status":"agent_timeout","configuration":{"agent":"Codex","agent_timeout_sec":7200,"coding_commit":null,"effort":"high","model":"xiaomi/mimo-v2.6-flash","provider":"OpenRouter","resources":null,"run_mode":"official_harbor_regrade","task_commit":"93c1f4e3d69dd992a651853520a7a2b707e7851a","verifier_commit":"93c1f4e3d69dd992a651853520a7a2b707e7851a","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-mimo-v2.6-flash-high","finished_at":"2026-10-28T22:57:20.711750Z","id":"reasoning-query-rewriting-mimo-v2.6-flash-high-808a15537ba5","redacted_fields":0,"reward":{"accuracy_at_3":0.1,"jailbreak_judge":1.0,"reward":0.1},"reward_sha256":"c5dc32a0e5e6c3329f4bd2748e4b9470306c958e92d654ad4a5ce5008bcf1235","score":10.0,"started_at":"2026-10-28T20:57:18.052204Z","steps":127,"task":"reasoning-query-rewriting","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":19763424,"input_tokens":27023822,"output_tokens":621378,"reasoning_output_tokens":6770,"response_count":123,"total_tokens":27645200},"trajectory_sha256":"2d7046045201ee5df1ebdf30fa1e72a2008db2ecfc25389b61f1fb5f27f5f4c7","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/reasoning-query-rewriting-mimo-v2.6-flash-high-808a15537ba5/manifest.json","valid":true,"cost_usd":1.2457791472},{"agent_seconds":2280.354684,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":10800.0,"coding_commit":null,"effort":"max","model":"deepseek/deepseek-v4.1-flash-1017","provider":"OpenRouter","resources":{"cpus":4,"gpus":0,"memory_mb":16384,"storage_mb":65536},"run_mode":"official_harbor_full","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"a1ede723abe0add54045301f49a187e11339c134","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-deepseek-v4.1-flash-1017-max","finished_at":"2026-10-24T10:14:06.461976Z","id":"rust-bm25-search-deepseek-v4.1-flash-1017-max-28890c17e54f","redacted_fields":0,"reward":0,"reward_sha256":"9a271f2a916b0b6ee6cecb2426f0b3206ef074578be55d9bc94f6f3fe3ab86aa","score":0.0,"started_at":"2026-10-24T09:36:06.107292Z","steps":153,"task":"rust-bm25-search","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":23552768,"input_tokens":25640463,"output_tokens":193323,"reasoning_output_tokens":68072,"response_count":150,"total_tokens":25833786},"trajectory_sha256":"fc08589812426a3cb8583d95b54385c5e67027e63c1e918d7ad837b23ed32421","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/rust-bm25-search-deepseek-v4.1-flash-1017-max-28890c17e54f/manifest.json","valid":true,"cost_usd":0.499806354},{"agent_seconds":1991.547174,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":10800.0,"coding_commit":null,"effort":"max","model":"openai/gpt-6-luna","provider":"OpenRouter","resources":{"cpus":4,"gpus":0,"memory_mb":16384,"storage_mb":65536},"run_mode":"official_harbor_full","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"a1ede723abe0add54045301f49a187e11339c134","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-gpt-6-luna-max","finished_at":"2026-10-24T08:30:52.265400Z","id":"rust-bm25-search-gpt-6-luna-max-639c7d699981","redacted_fields":0,"reward":0.240572649101,"reward_sha256":"45385d15366449a91599be3e4de739ee444eb657e676778f2ebfce3effec1cb1","score":24.057264910100002,"started_at":"2026-10-24T07:57:40.718226Z","steps":65,"task":"rust-bm25-search","tokens":{"cache_write_input_tokens":166186,"cached_input_tokens":7018201,"input_tokens":7251291,"output_tokens":128589,"reasoning_output_tokens":99404,"response_count":62,"total_tokens":7379880},"trajectory_sha256":"44b3d2cc63b92fef9446f81840248eda53ba1004e737f1076227c9c05c4eb734","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/rust-bm25-search-gpt-6-luna-max-639c7d699981/manifest.json","valid":true,"cost_usd":0.16194016},{"agent_seconds":1074.784631,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":10800.0,"coding_commit":null,"effort":"high","model":"xiaomi/mimo-v2.6-flash","provider":"OpenRouter","resources":{"cpus":4,"gpus":0,"memory_mb":16384,"storage_mb":65536},"run_mode":"official_harbor_full","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"a1ede723abe0add54045301f49a187e11339c134","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-mimo-v2.6-flash-high","finished_at":"2026-10-24T08:15:34.846830Z","id":"rust-bm25-search-mimo-v2.6-flash-high-20c9748b5206","redacted_fields":0,"reward":0,"reward_sha256":"9a271f2a916b0b6ee6cecb2426f0b3206ef074578be55d9bc94f6f3fe3ab86aa","score":0.0,"started_at":"2026-10-24T07:57:40.062199Z","steps":64,"task":"rust-bm25-search","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":5732736,"input_tokens":6010975,"output_tokens":50346,"reasoning_output_tokens":15173,"response_count":61,"total_tokens":6061321},"trajectory_sha256":"22d1ad4aea257b2ecdeb7b8513f090dce16230662dce59631dc3da2d0c3bc964","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/rust-bm25-search-mimo-v2.6-flash-high-20c9748b5206/manifest.json","valid":true,"cost_usd":0.0691020008},{"agent_seconds":7203.460206,"agent_status":"agent_budget_exhausted","configuration":{"agent":"Codex","agent_timeout_sec":7200,"coding_commit":null,"effort":"max","model":"deepseek/deepseek-v4.1-flash-1017","provider":"OpenRouter","resources":{"cpus":8,"gpus":1,"memory_mb":32768,"storage_mb":102400},"run_mode":"official_harbor_full","task_commit":null,"verifier_commit":"1c433c71bd7b741f34924d4414e0b772354e3645","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-deepseek-v4.1-flash-1017-max","finished_at":"2026-10-27T19:58:01.815402Z","id":"same-place-image-search-deepseek-v4.1-flash-1017-max-e281375ddf5a","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"reward":0.5201110939,"same_place_ndcg":0.7382804349},"reward_sha256":"7aab6ab21c4d288d2ae3da7a4a3e4e2d73505ecd3a43bcfa6677f3e80939bf3a","score":52.01110939,"started_at":"2026-10-27T17:57:58.355196Z","steps":364,"task":"same-place-image-search","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":37245891,"input_tokens":37915500,"output_tokens":205066,"reasoning_output_tokens":140932,"response_count":360,"total_tokens":38120566},"trajectory_sha256":"bd763e3bf44d1d0b64c7bccef32781d8e115b7d1f3fc46481e6527feaf04e51b","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/same-place-image-search-deepseek-v4.1-flash-1017-max-e281375ddf5a/manifest.json","valid":true,"cost_usd":0.335218623},{"agent_seconds":3765.52067,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":7200,"coding_commit":null,"effort":"max","model":"openai/gpt-6-luna","provider":"OpenRouter","resources":{"cpus":8,"gpus":1,"memory_mb":32768,"storage_mb":102400},"run_mode":"official_harbor_full","task_commit":null,"verifier_commit":"1c433c71bd7b741f34924d4414e0b772354e3645","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-gpt-6-luna-max","finished_at":"2026-10-27T19:00:43.871691Z","id":"same-place-image-search-gpt-6-luna-max-f29df079f279","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"reward":0.5756103002,"same_place_ndcg":0.7685483322},"reward_sha256":"0fc9c5125051ab0a9b68efc6435d7e33c33ec0ed37dc1496ba29213f1b1ff932","score":57.561030020000004,"started_at":"2026-10-27T17:57:58.351021Z","steps":89,"task":"same-place-image-search","tokens":{"cache_write_input_tokens":164972,"cached_input_tokens":9131007,"input_tokens":9296237,"output_tokens":113042,"reasoning_output_tokens":79003,"response_count":86,"total_tokens":9409279},"trajectory_sha256":"f435e3a1abe40ea8e648f1511956f9e18c94583f1c42981f7d3d456213b8f85f","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/same-place-image-search-gpt-6-luna-max-f29df079f279/manifest.json","valid":true,"cost_usd":0.16847837},{"agent_seconds":517.111408,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":7200,"coding_commit":null,"effort":"high","model":"xiaomi/mimo-v2.6-flash","provider":"OpenRouter","resources":{"cpus":8,"gpus":1,"memory_mb":32768,"storage_mb":102400},"run_mode":"official_harbor_full","task_commit":null,"verifier_commit":"1c433c71bd7b741f34924d4414e0b772354e3645","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-mimo-v2.6-flash-high","finished_at":"2026-10-27T18:06:35.432352Z","id":"same-place-image-search-mimo-v2.6-flash-high-597c6d564c37","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"reward":0.0,"same_place_ndcg":0.0},"reward_sha256":"623ac7add0c1bf473ff00814a13f673cc6d2f9c8662cc1aee1cf0ec2d9b00462","score":0.0,"started_at":"2026-10-27T17:57:58.320944Z","steps":26,"task":"same-place-image-search","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":698880,"input_tokens":743896,"output_tokens":11141,"reasoning_output_tokens":562,"response_count":24,"total_tokens":755037},"trajectory_sha256":"19efd8d6e3bdc5ef0e66501cda7d767f86e26ea8695eae85bfb97ced39cb4722","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/same-place-image-search-mimo-v2.6-flash-high-597c6d564c37/manifest.json","valid":true,"cost_usd":0.011378584},{"agent_seconds":7200.513201,"agent_status":"agent_timeout","configuration":{"agent":"Codex","agent_timeout_sec":7200.0,"coding_commit":null,"effort":"max","model":"deepseek/deepseek-v4.1-flash-1017","provider":"OpenRouter","resources":null,"run_mode":"official_harbor_regrade","task_commit":null,"verifier_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-deepseek-v4.1-flash-1017-max","finished_at":"2026-10-23T22:19:50.710160Z","id":"scientific-paper-qa-deepseek-v4.1-flash-1017-max-422713068cf0","redacted_fields":0,"reward":{"evidence_accuracy":0.0,"jailbreak_judge":1.0,"llm_judge_accuracy":0.0,"reward":0.0},"reward_sha256":"b6531f02ab70c154668ccc55b6977e1e1564af0cde0443a28ec959b147ac0601","score":0.0,"started_at":"2026-10-23T20:19:50.196959Z","steps":367,"task":"scientific-paper-qa","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":79240576,"input_tokens":80020322,"output_tokens":221219,"reasoning_output_tokens":82819,"response_count":364,"total_tokens":80241541},"trajectory_sha256":"38d680a9f2dc801381853fae9f2ca4d2bb8644f5973ae9e49a029b9e64c188b3","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/scientific-paper-qa-deepseek-v4.1-flash-1017-max-422713068cf0/manifest.json","valid":true,"cost_usd":0.487415028},{"agent_seconds":7200.463793,"agent_status":"agent_timeout","configuration":{"agent":"Codex","agent_timeout_sec":7200.0,"coding_commit":null,"effort":"max","model":"openai/gpt-6-luna","provider":"OpenRouter","resources":null,"run_mode":"full_agent_and_verifier","task_commit":null,"verifier_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-gpt-6-luna-max","finished_at":"2026-10-24T00:59:22.631458Z","id":"scientific-paper-qa-gpt-6-luna-max-e3858ea7d08e","redacted_fields":0,"reward":{"evidence_accuracy":0.7,"jailbreak_judge":1.0,"llm_judge_accuracy":0.2,"reward":0.2},"reward_sha256":"75751f957832f539c76e349022e450189abea9392645d3492610b73e7f1b8618","score":20.0,"started_at":"2026-10-23T22:59:22.167665Z","steps":260,"task":"scientific-paper-qa","tokens":{"cache_write_input_tokens":631218,"cached_input_tokens":57333274,"input_tokens":58004541,"output_tokens":237577,"reasoning_output_tokens":172927,"response_count":256,"total_tokens":58242118},"trajectory_sha256":"26c17c19c7287265028fab02e62feaa66fa81067757db995d0b77fb4aef3d098","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/scientific-paper-qa-gpt-6-luna-max-e3858ea7d08e/manifest.json","valid":true,"cost_usd":1.13289106},{"agent_seconds":7200.45021,"agent_status":"agent_timeout","configuration":{"agent":"Codex","agent_timeout_sec":7200,"coding_commit":null,"effort":"high","model":"xiaomi/mimo-v2.6-flash","provider":"OpenRouter","resources":null,"run_mode":"official_harbor_regrade","task_commit":null,"verifier_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-mimo-v2.6-flash-high","finished_at":"2026-10-23T22:56:22.431995Z","id":"scientific-paper-qa-mimo-v2.6-flash-high-3f2e72195a73","redacted_fields":0,"reward":{"evidence_accuracy":0.4,"jailbreak_judge":1.0,"llm_judge_accuracy":0.2,"reward":0.2},"reward_sha256":"e98935c4c44a75de37809840f21def21a8fdb5a3f96222cb9c5f071c6c96a6bf","score":20.0,"started_at":"2026-10-23T20:56:21.981785Z","steps":144,"task":"scientific-paper-qa","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":11906816,"input_tokens":16093014,"output_tokens":592360,"reasoning_output_tokens":5513,"response_count":141,"total_tokens":16685374},"trajectory_sha256":"18d40bc327dd98ec28e896dc6d0d354cd6640e71a5b5da7febad40d5792445bf","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/scientific-paper-qa-mimo-v2.6-flash-high-3f2e72195a73/manifest.json","valid":true,"cost_usd":0.7852676048},{"agent_seconds":21603.475565,"agent_status":"agent_timeout","configuration":{"agent":"Codex","agent_timeout_sec":21600.0,"coding_commit":null,"effort":"max","model":"deepseek/deepseek-v4.1-flash-1017","provider":"OpenRouter","resources":{"cpus":8,"gpus":1,"memory_mb":32768,"storage_mb":102400},"run_mode":"official_harbor_full","task_commit":"052045836520335c1caa5e4a06ed887faacef55f","verifier_commit":"052045836520335c1caa5e4a06ed887faacef55f","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-deepseek-v4.1-flash-1017-max","finished_at":"2026-10-25T13:51:09.150357Z","id":"sketch-text-search-deepseek-v4.1-flash-1017-max-f19fc4a6d4e9","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"reward":0.0,"sketch_text_quality":0.0},"reward_sha256":"daff9edbd7fffe2afdf6447024411a4543ddbd72265b194d36c65dd48fa91466","score":0.0,"started_at":"2026-10-25T07:51:05.674792Z","steps":204,"task":"sketch-text-search","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":18204346,"input_tokens":18601297,"output_tokens":155389,"reasoning_output_tokens":78971,"response_count":200,"total_tokens":18756686},"trajectory_sha256":"48d33ef749b9ec4eb438529251064d31e517e537a623b5dec7920c17d032d343","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/sketch-text-search-deepseek-v4.1-flash-1017-max-f19fc4a6d4e9/manifest.json","valid":true,"cost_usd":0.207389088},{"agent_seconds":5041.757159,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":21600.0,"coding_commit":null,"effort":"max","model":"openai/gpt-6-luna","provider":"OpenRouter","resources":{"cpus":8,"gpus":1,"memory_mb":32768,"storage_mb":102400},"run_mode":"official_harbor_full","task_commit":"052045836520335c1caa5e4a06ed887faacef55f","verifier_commit":"052045836520335c1caa5e4a06ed887faacef55f","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-gpt-6-luna-max","finished_at":"2026-10-25T12:11:59.020394Z","id":"sketch-text-search-gpt-6-luna-max-d258a9625558","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"reward":0.0349844237,"sketch_text_quality":0.0753488996},"reward_sha256":"90b657adc6b7aa553d9f8d6fd4a8134a93480a577b5dc140d35df86a7cdaf34f","score":3.4984423700000002,"started_at":"2026-10-25T10:47:57.263235Z","steps":140,"task":"sketch-text-search","tokens":{"cache_write_input_tokens":256508,"cached_input_tokens":20783360,"input_tokens":21040279,"output_tokens":157333,"reasoning_output_tokens":102468,"response_count":137,"total_tokens":21197612},"trajectory_sha256":"a28ff24b0939b4fc1c82ee58612b3d59b7e32aafc6deef254d95c5cd9ecd234b","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/sketch-text-search-gpt-6-luna-max-d258a9625558/manifest.json","valid":true,"cost_usd":0.3186047},{"agent_seconds":6765.98479,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":21600.0,"coding_commit":null,"effort":"high","model":"xiaomi/mimo-v2.6-flash","provider":"OpenRouter","resources":{"cpus":8,"gpus":1,"memory_mb":32768,"storage_mb":102400},"run_mode":"official_harbor_regrade","task_commit":"052045836520335c1caa5e4a06ed887faacef55f","verifier_commit":"052045836520335c1caa5e4a06ed887faacef55f","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-mimo-v2.6-flash-high","finished_at":"2026-10-25T11:07:46.704131Z","id":"sketch-text-search-mimo-v2.6-flash-high-a6d02b1f0a15","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"reward":0.0137383178,"sketch_text_quality":0.029589372},"reward_sha256":"6fda1d666b4557198357e8e804b7ebff59ce5491b01663c2b2b40f57caf8b28f","score":1.37383178,"started_at":"2026-10-25T09:15:00.719341Z","steps":175,"task":"sketch-text-search","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":19133632,"input_tokens":20011474,"output_tokens":88443,"reasoning_output_tokens":13425,"response_count":172,"total_tokens":20099917},"trajectory_sha256":"6bacd10754a3fb121a8871104931fedd5a9dd4343aa315c8ed1091b13460abd2","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/sketch-text-search-mimo-v2.6-flash-high-a6d02b1f0a15/manifest.json","valid":true,"cost_usd":0.2012360896},{"agent_seconds":7179.989894,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":7200.0,"coding_commit":null,"effort":"max","model":"deepseek/deepseek-v4.1-flash-1017","provider":"OpenRouter","resources":{"cpus":1,"gpus":0,"memory_mb":32768,"storage_mb":81920},"run_mode":"official_harbor_full","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"a1ede723abe0add54045301f49a187e11339c134","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-deepseek-v4.1-flash-1017-max","finished_at":"2026-10-24T15:52:31.508947Z","id":"sparse-retrieval-pruning-deepseek-v4.1-flash-1017-max-786e50a1a9a5","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"retrieval_quality_and_latency":0.0,"reward":0.0,"score":0.0},"reward_sha256":"02b20b84040e4f78b91eefe0ffe40d8348e8a557581da5f2de1f6f31d108dec0","score":0.0,"started_at":"2026-10-24T13:52:51.519053Z","steps":504,"task":"sparse-retrieval-pruning","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":88309151,"input_tokens":89879717,"output_tokens":326595,"reasoning_output_tokens":170989,"response_count":501,"total_tokens":90206312},"trajectory_sha256":"38b1624730f69eaddf7b538bdfd2416d30302e203b7b9923bfb5e111690ab928","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/sparse-retrieval-pruning-deepseek-v4.1-flash-1017-max-786e50a1a9a5/manifest.json","valid":true,"cost_usd":0.696469353},{"agent_seconds":2976.557339,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":7200.0,"coding_commit":null,"effort":"max","model":"openai/gpt-6-luna","provider":"OpenRouter","resources":{"cpus":1,"gpus":0,"memory_mb":32768,"storage_mb":81920},"run_mode":"official_harbor_full","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"a1ede723abe0add54045301f49a187e11339c134","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-gpt-6-luna-max","finished_at":"2026-10-24T14:42:28.826035Z","id":"sparse-retrieval-pruning-gpt-6-luna-max-98de8be9fd73","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"retrieval_quality_and_latency":0.0,"reward":0.0,"score":0.0},"reward_sha256":"02b20b84040e4f78b91eefe0ffe40d8348e8a557581da5f2de1f6f31d108dec0","score":0.0,"started_at":"2026-10-24T13:52:52.268696Z","steps":108,"task":"sparse-retrieval-pruning","tokens":{"cache_write_input_tokens":227400,"cached_input_tokens":14255623,"input_tokens":14483338,"output_tokens":177938,"reasoning_output_tokens":139405,"response_count":105,"total_tokens":14661276},"trajectory_sha256":"f22a4a3e11ea48b808fba0f126c0d94303eb75336f488b25b0966925f797a69f","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/sparse-retrieval-pruning-gpt-6-luna-max-98de8be9fd73/manifest.json","valid":true,"cost_usd":0.25998173},{"agent_seconds":3322.7061,"agent_status":"output_limit","configuration":{"agent":"Codex","agent_timeout_sec":7200.0,"coding_commit":null,"effort":"high","model":"xiaomi/mimo-v2.6-flash","provider":"OpenRouter","resources":{"cpus":1,"gpus":0,"memory_mb":32768,"storage_mb":81920},"run_mode":"official_harbor_full","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"a1ede723abe0add54045301f49a187e11339c134","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-mimo-v2.6-flash-high","finished_at":"2026-10-24T14:48:13.919893Z","id":"sparse-retrieval-pruning-mimo-v2.6-flash-high-e0272e827772","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"retrieval_quality_and_latency":0.0,"reward":0.0,"score":0.0},"reward_sha256":"02b20b84040e4f78b91eefe0ffe40d8348e8a557581da5f2de1f6f31d108dec0","score":0.0,"started_at":"2026-10-24T13:52:51.213793Z","steps":37,"task":"sparse-retrieval-pruning","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":2453150,"input_tokens":3900004,"output_tokens":532904,"reasoning_output_tokens":3688,"response_count":33,"total_tokens":4432908},"trajectory_sha256":"bd9215b86a099f108c3f88522fe96591d0cab4c3094b2cbf6f3bb2523813ced8","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/sparse-retrieval-pruning-mimo-v2.6-flash-high-e0272e827772/manifest.json","valid":true,"cost_usd":0.3586415},{"agent_seconds":7201.790314,"agent_status":"agent_timeout","configuration":{"agent":"Codex","agent_timeout_sec":7200.0,"coding_commit":null,"effort":"max","model":"deepseek/deepseek-v4.1-flash-1017","provider":"OpenRouter","resources":{"cpus":8,"gpus":1,"memory_mb":32768,"storage_mb":102400},"run_mode":"official_harbor_full","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"bc2ef55cf5bc4e5133c32816a7874368192a2f1e","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-deepseek-v4.1-flash-1017-max","finished_at":"2026-10-24T23:57:27.781925Z","id":"spoken-query-search-deepseek-v4.1-flash-1017-max-9ac8dcb8fcc6","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"reward":0.1433490522,"spoken_macro_ndcg":0.4329231796},"reward_sha256":"3de467fbb6947402774185fdf56500878d9dac9c05c7b4d45f2be4a960c39c69","score":14.33490522,"started_at":"2026-10-24T21:57:25.991611Z","steps":286,"task":"spoken-query-search","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":44942065,"input_tokens":45669931,"output_tokens":355065,"reasoning_output_tokens":219448,"response_count":282,"total_tokens":46024996},"trajectory_sha256":"62e80d15c0496500e43956abde6031bab27508f84f7a9101942424e943905b1f","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/spoken-query-search-deepseek-v4.1-flash-1017-max-9ac8dcb8fcc6/manifest.json","valid":true,"cost_usd":0.457045095},{"agent_seconds":7201.817569,"agent_status":"agent_timeout","configuration":{"agent":"Codex","agent_timeout_sec":7200.0,"coding_commit":null,"effort":"max","model":"openai/gpt-6-luna","provider":"OpenRouter","resources":{"cpus":8,"gpus":1,"memory_mb":32768,"storage_mb":102400},"run_mode":"official_harbor_full","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"bc2ef55cf5bc4e5133c32816a7874368192a2f1e","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-gpt-6-luna-max","finished_at":"2026-10-25T01:29:57.022638Z","id":"spoken-query-search-gpt-6-luna-max-7f985f759ae3","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"reward":0.1382760164,"spoken_macro_ndcg":0.4176022916},"reward_sha256":"f92f4f867dc8eadef964ae55dfaf763f03149468693e61cf1cd9b29b6158ca23","score":13.827601640000001,"started_at":"2026-10-24T23:29:55.205069Z","steps":117,"task":"spoken-query-search","tokens":{"cache_write_input_tokens":486535,"cached_input_tokens":18426150,"input_tokens":18913018,"output_tokens":190681,"reasoning_output_tokens":156675,"response_count":111,"total_tokens":19103699},"trajectory_sha256":"eeaf090182e1c14deb3a4529a78ba02e2f94655706baa1ca6328cc23b2427aed","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/spoken-query-search-gpt-6-luna-max-7f985f759ae3/manifest.json","valid":true,"cost_usd":0.340452175},{"agent_seconds":7201.837063,"agent_status":"agent_timeout","configuration":{"agent":"Codex","agent_timeout_sec":7200.0,"coding_commit":null,"effort":"high","model":"xiaomi/mimo-v2.6-flash","provider":"OpenRouter","resources":{"cpus":8,"gpus":1,"memory_mb":32768,"storage_mb":102400},"run_mode":"official_harbor_full","task_commit":"bb34e915be5c2b2948146c2d7862cb58652d6300","verifier_commit":"bc2ef55cf5bc4e5133c32816a7874368192a2f1e","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-mimo-v2.6-flash-high","finished_at":"2026-10-24T21:52:09.429527Z","id":"spoken-query-search-mimo-v2.6-flash-high-37353bb5d4de","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"reward":0.0197176962,"spoken_macro_ndcg":0.0595486863},"reward_sha256":"ccd01ae359602f3fa8c80a7272a4c8856aa65f269d9de35ad1e100e1ec4e3cf5","score":1.97176962,"started_at":"2026-10-24T19:52:07.592464Z","steps":207,"task":"spoken-query-search","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":22962322,"input_tokens":25674079,"output_tokens":534300,"reasoning_output_tokens":72480,"response_count":203,"total_tokens":26208379},"trajectory_sha256":"d6b83a66a7b8e9baf54a3360708ecf3a9e3f82766584466dbe4a09d18bb9cd19","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/spoken-query-search-mimo-v2.6-flash-high-37353bb5d4de/manifest.json","valid":true,"cost_usd":0.5935444816},{"agent_seconds":10269.835375,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":14400.0,"coding_commit":null,"effort":"max","model":"deepseek/deepseek-v4.1-flash-1017","provider":"OpenRouter","resources":null,"run_mode":"official_harbor_regrade","task_commit":null,"verifier_commit":"8287076ba397b9266c9a399a857a89ba4815e2cb","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-deepseek-v4.1-flash-1017-max","finished_at":"2026-10-23T21:41:25.538174Z","id":"training-error-tracing-deepseek-v4.1-flash-1017-max-2419e1ce895c","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"reward":0.0,"tracing_improvement":0.0},"reward_sha256":"4c105b06cff24c88ff19850fa179d69e9459fbc88285320493c27ef47952d2ba","score":0.0,"started_at":"2026-10-23T18:50:15.702799Z","steps":491,"task":"training-error-tracing","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":99769088,"input_tokens":103296442,"output_tokens":330365,"reasoning_output_tokens":213383,"response_count":488,"total_tokens":103626807},"trajectory_sha256":"ee524771980cbb7f01c3f3ef212525bb5aaa5851cb5f034e3606a8d1fa5c61e7","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/training-error-tracing-deepseek-v4.1-flash-1017-max-2419e1ce895c/manifest.json","valid":true,"cost_usd":1.026629364},{"agent_seconds":1495.81157,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":14400.0,"coding_commit":null,"effort":"max","model":"openai/gpt-6-luna","provider":"OpenRouter","resources":null,"run_mode":"full_agent_and_verifier","task_commit":null,"verifier_commit":"8287076ba397b9266c9a399a857a89ba4815e2cb","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-gpt-6-luna-max","finished_at":"2026-10-23T23:39:58.034049Z","id":"training-error-tracing-gpt-6-luna-max-551ca02ceb47","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"reward":0.0,"tracing_improvement":0.0},"reward_sha256":"4c105b06cff24c88ff19850fa179d69e9459fbc88285320493c27ef47952d2ba","score":0.0,"started_at":"2026-10-23T23:15:02.222479Z","steps":74,"task":"training-error-tracing","tokens":{"cache_write_input_tokens":167142,"cached_input_tokens":7356046,"input_tokens":7523401,"output_tokens":93179,"reasoning_output_tokens":74425,"response_count":71,"total_tokens":7616580},"trajectory_sha256":"f1c93f4b24afef3946000b72b11eeef742193073417a1402fdb9718231307f1d","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/training-error-tracing-gpt-6-luna-max-551ca02ceb47/manifest.json","valid":true,"cost_usd":0.14106401},{"agent_seconds":2411.734171,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":14400.0,"coding_commit":null,"effort":"high","model":"xiaomi/mimo-v2.6-flash","provider":"OpenRouter","resources":null,"run_mode":"full_agent_and_verifier","task_commit":null,"verifier_commit":"8287076ba397b9266c9a399a857a89ba4815e2cb","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-mimo-v2.6-flash-high","finished_at":"2026-10-23T23:38:45.220824Z","id":"training-error-tracing-mimo-v2.6-flash-high-e3fd1694f69a","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"reward":0.2,"tracing_improvement":0.2},"reward_sha256":"07a01c46129633de14f59ca800f0546a2711888fa8d401e6bf15411f90ed59a0","score":20.0,"started_at":"2026-10-23T22:58:33.486653Z","steps":203,"task":"training-error-tracing","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":17373891,"input_tokens":18824905,"output_tokens":56478,"reasoning_output_tokens":13603,"response_count":200,"total_tokens":18881383},"trajectory_sha256":"bbcdf7259e96b320568c473bf09df3407dfa124c4f115f183e807302ff52cfe4","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/training-error-tracing-mimo-v2.6-flash-high-e3fd1694f69a/manifest.json","valid":true,"cost_usd":0.2676026948},{"agent_seconds":59.484717,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":21600.0,"coding_commit":null,"effort":"max","model":"deepseek/deepseek-v4.1-flash-1017","provider":"OpenRouter","resources":{"cpus":8,"gpus":1,"memory_mb":32768,"storage_mb":102400},"run_mode":"official_harbor_full","task_commit":"052045836520335c1caa5e4a06ed887faacef55f","verifier_commit":"052045836520335c1caa5e4a06ed887faacef55f","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-deepseek-v4.1-flash-1017-max","finished_at":"2026-10-25T09:15:57.413125Z","id":"video-moment-search-deepseek-v4.1-flash-1017-max-18c138d537f8","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"reward":0.0,"video_grid_utility":0.0},"reward_sha256":"20b13cc74c07c5bdbaf7231d99cf4c78654d749a3328e4068dc036f09a31e5fe","score":0.0,"started_at":"2026-10-25T09:14:57.928408Z","steps":7,"task":"video-moment-search","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":33920,"input_tokens":65808,"output_tokens":7592,"reasoning_output_tokens":2627,"response_count":4,"total_tokens":73400},"trajectory_sha256":"cbc560ee949478abe5c0e61fdec980c00435d27c87155a91850d4dcb3cbed23f","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/video-moment-search-deepseek-v4.1-flash-1017-max-18c138d537f8/manifest.json","valid":true,"cost_usd":0.00944016},{"agent_seconds":1468.323094,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":21600.0,"coding_commit":null,"effort":"max","model":"openai/gpt-6-luna","provider":"OpenRouter","resources":{"cpus":8,"gpus":1,"memory_mb":32768,"storage_mb":102400},"run_mode":"official_harbor_full","task_commit":"052045836520335c1caa5e4a06ed887faacef55f","verifier_commit":"052045836520335c1caa5e4a06ed887faacef55f","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-gpt-6-luna-max","finished_at":"2026-10-25T12:48:12.363482Z","id":"video-moment-search-gpt-6-luna-max-cbab874afc52","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"reward":0.227628793,"video_grid_utility":0.2248715024},"reward_sha256":"112c6db9cec11f913d02725136064102d8c23d54540abd5e61b2e02b7e6e7703","score":22.762879299999998,"started_at":"2026-10-25T12:23:44.040388Z","steps":44,"task":"video-moment-search","tokens":{"cache_write_input_tokens":152434,"cached_input_tokens":3622157,"input_tokens":3774714,"output_tokens":108385,"reasoning_output_tokens":69135,"response_count":41,"total_tokens":3883099},"trajectory_sha256":"6842a151dfb45fca777fb5c621f96f21e5af3c7ef9c261bf960073415d30515d","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/video-moment-search-gpt-6-luna-max-cbab874afc52/manifest.json","valid":true,"cost_usd":0.10948062},{"agent_seconds":1011.808312,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":21600.0,"coding_commit":null,"effort":"high","model":"xiaomi/mimo-v2.6-flash","provider":"OpenRouter","resources":{"cpus":8,"gpus":1,"memory_mb":32768,"storage_mb":102400},"run_mode":"official_harbor_regrade","task_commit":"052045836520335c1caa5e4a06ed887faacef55f","verifier_commit":"052045836520335c1caa5e4a06ed887faacef55f","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-mimo-v2.6-flash-high","finished_at":"2026-10-25T09:37:21.188628Z","id":"video-moment-search-mimo-v2.6-flash-high-ab172ae1fde0","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"reward":0.1086851068,"video_grid_utility":0.1064864665},"reward_sha256":"2432a4754772e9b4c5a0273310e5bbd9e1cb2365cdfd40b84e03efaf56f12d25","score":10.86851068,"started_at":"2026-10-25T09:20:29.380316Z","steps":54,"task":"video-moment-search","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":2585152,"input_tokens":2668851,"output_tokens":30250,"reasoning_output_tokens":3438,"response_count":51,"total_tokens":2699101},"trajectory_sha256":"65928f8212bdf9ac7d25ae0b3189c0872aa38edb9a05a34176d2d0dff46acc27","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/video-moment-search-mimo-v2.6-flash-high-ab172ae1fde0/manifest.json","valid":true,"cost_usd":0.0274262856},{"agent_seconds":6918.79331,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":null,"coding_commit":null,"effort":"max","model":"deepseek/deepseek-v4.1-flash-1017","provider":"OpenRouter","resources":null,"run_mode":"saved_answers_uniform_full_judge_regrade","task_commit":"ddb979fcd5d562af55a3d6e0d8a2b7a961770433","verifier_commit":"ddb979fcd5d562af55a3d6e0d8a2b7a961770433","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-deepseek-v4.1-flash-1017-max","finished_at":"2026-10-25T01:41:01.548976Z","id":"visual-document-qa-deepseek-v4.1-flash-1017-max-1c98bbda6bb2","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"reward":0.2165577076,"visual_document_qa":0.4467908105},"reward_sha256":"7a5a68acbeb059052cb78d139feadedf716888bff3f9955ec7c7e3f621107a7b","score":21.65577076,"started_at":"2026-10-24T23:45:42.755666Z","steps":216,"task":"visual-document-qa","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":39592146,"input_tokens":43405107,"output_tokens":375545,"reasoning_output_tokens":229795,"response_count":212,"total_tokens":43780652},"trajectory_sha256":"80ca22c0b90a3caa18bbaa421df190145a362861b036cdc72fed513726986e37","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/visual-document-qa-deepseek-v4.1-flash-1017-max-1c98bbda6bb2/manifest.json","valid":true,"cost_usd":0.916047588},{"agent_seconds":2920.196329,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":null,"coding_commit":null,"effort":"max","model":"openai/gpt-6-luna","provider":"OpenRouter","resources":null,"run_mode":"saved_answers_uniform_full_judge_regrade","task_commit":"ddb979fcd5d562af55a3d6e0d8a2b7a961770433","verifier_commit":"ddb979fcd5d562af55a3d6e0d8a2b7a961770433","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-gpt-6-luna-max","finished_at":"2026-10-25T00:57:40.182692Z","id":"visual-document-qa-gpt-6-luna-max-6ddef3fe3d09","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"reward":0.1779622894,"visual_document_qa":0.3871560513},"reward_sha256":"02af5b7993bf1b9c30183c4871407e9318ca0b3164485c4c4f356381c8509616","score":17.79622894,"started_at":"2026-10-25T00:08:59.986363Z","steps":113,"task":"visual-document-qa","tokens":{"cache_write_input_tokens":348315,"cached_input_tokens":16725088,"input_tokens":17102874,"output_tokens":189670,"reasoning_output_tokens":141090,"response_count":110,"total_tokens":17292544},"trajectory_sha256":"bd2bbe8f4baada2d6db209da2db5c74684d094dc85cfb88242c9e206d09836f3","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/visual-document-qa-gpt-6-luna-max-6ddef3fe3d09/manifest.json","valid":true,"cost_usd":0.308572355},{"agent_seconds":1434.914741,"agent_status":"normal","configuration":{"agent":"Codex","agent_timeout_sec":null,"coding_commit":null,"effort":"high","model":"xiaomi/mimo-v2.6-flash","provider":"OpenRouter","resources":null,"run_mode":"saved_answers_uniform_full_judge_regrade","task_commit":"ddb979fcd5d562af55a3d6e0d8a2b7a961770433","verifier_commit":"ddb979fcd5d562af55a3d6e0d8a2b7a961770433","version":"0.164.0"},"entry":"2026-10-31-local-1233340bbd-mimo-v2.6-flash-high","finished_at":"2026-10-25T00:26:39.166914Z","id":"visual-document-qa-mimo-v2.6-flash-high-009552f63a14","redacted_fields":0,"reward":{"jailbreak_judge":1.0,"reward":0.067726823,"visual_document_qa":0.1473393574},"reward_sha256":"2af54891236871755d3b2eeb909c4d42ab55669e97d68e2b24c4cc606d9674b7","score":6.7726823000000005,"started_at":"2026-10-25T00:02:44.252173Z","steps":115,"task":"visual-document-qa","tokens":{"cache_write_input_tokens":0,"cached_input_tokens":6460608,"input_tokens":6722315,"output_tokens":31152,"reasoning_output_tokens":2753,"response_count":113,"total_tokens":6753467},"trajectory_sha256":"1d2a9db3d53cdbcb5a22fb1cd8844a20d19b017b22c89135a75ef9e0e3df1039","trajectory_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/trials/visual-document-qa-mimo-v2.6-flash-high-009552f63a14/manifest.json","valid":true,"cost_usd":0.0634512424}],"data_source":{"provider":"huggingface","base_url":"https://huggingface.co/datasets/search-swe/trajectories/resolve/824b6d863b69b1f86b9d73b365dca28af200de7c/releases/2026-10-30-3090361873/"},"download_url":"/leaderboard/results.json","cost_methodology":{"type":"estimated","schema":1,"as_of":"2026-11-01","currency":"USD","unit":"USD per million tokens","policy":"Standard real-time API list prices; fixed DeepSeek off-peak rates for all requests, Xiaomi overseas USD rates. No time-of-day switching, batch, account, or regional adjustments.","scope":"Selected coding agent model requests only; excludes verifier, task-internal APIs, infrastructure and unrelated runs. Recorded cache reads and writes are subsets of input. Reasoning is already included in output.","models":{"openai/gpt-6-luna":{"official_model":"gpt-6-luna","source":"https://developers.openai.com/api/docs/models/gpt-6-luna","rates":{"input":"0.10","cached_input":"0.01","cache_write_input":"0.125","output":"0.50"},"long_context":{"above_input_tokens":272000,"rates":{"input":"0.20","cached_input":"0.02","cache_write_input":"0.25","output":"0.75"}},"note":"Standard processing. Prompts above 272,000 input tokens use long-context rates for the entire request. Cache writes use the standard automatic-cache write rate."},"deepseek/deepseek-v4.1-flash-1017":{"official_model":"deepseek-flash (DeepSeek-V4.1-Flash-1017)","source":"https://api-docs.deepseek.com/quick_start/pricing/","rates":{"input":"0.15","cached_input":"0.003","cache_write_input":"0.15","output":"0.60"},"note":"Fixed official off-peak rates for every request, independent of run time. Cache misses, including writes, are charged at the input rate without a separate write surcharge."},"xiaomi/mimo-v2.6-flash":{"official_model":"mimo-v2.6-flash","source":"https://mimo.mi.com/docs/en-US/price/pay-as-you-go","rates":{"input":"0.14","cached_input":"0.0028","cache_write_input":"0.14","output":"0.28"},"note":"Overseas real-time API prices in USD. Cache writes have no additional charge; their input processing is charged at the cache-miss rate."}},"usage_sha256":"883b4a876a873693babc78091c580073f80642c157bd9a59efe40e4a9a201a55","pricing_sha256":"554c592135c62631c62f58a58329cda66835e7bdbe38409137c62845d2a2d9e8"}}
